Token | in |
Token ID | 287 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
glad | +0.018 |
pleased | +0.017 |
impressed | +0.016 |
sorry | +0.016 |
lucky | +0.016 |
Neg logit contributions | |
bons | -0.017 |
talk | -0.014 |
growth | -0.014 |
bars | -0.014 |
ales | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
glad | +0.005 |
pleased | +0.004 |
impressed | +0.004 |
okay | +0.004 |
sorry | +0.004 |
Neg logit contributions | |
bons | -0.005 |
growth | -0.004 |
talk | -0.004 |
bars | -0.004 |
shirts | -0.004 |
Token | bad |
Token ID | 2089 |
Feature activation | +9.49e-01 |
Pos logit contributions | |
glad | +0.100 |
pleased | +0.092 |
impressed | +0.086 |
okay | +0.085 |
sorry | +0.081 |
Neg logit contributions | |
bons | -0.120 |
bars | -0.115 |
growth | -0.112 |
talk | -0.109 |
bank | -0.103 |
Token | way |
Token ID | 835 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
glad | +0.058 |
pleased | +0.056 |
impressed | +0.052 |
okay | +0.050 |
sorry | +0.049 |
Neg logit contributions | |
bons | -0.048 |
growth | -0.042 |
talk | -0.041 |
bars | -0.040 |
ales | -0.037 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
glad | +0.146 |
pleased | +0.143 |
impressed | +0.135 |
sorry | +0.131 |
lucky | +0.129 |
Neg logit contributions | |
bons | -0.087 |
bars | -0.078 |
talk | -0.077 |
growth | -0.071 |
ales | -0.071 |
Token | Once |
Token ID | 7454 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
glad | +0.014 |
pleased | +0.014 |
impressed | +0.012 |
okay | +0.012 |
sorry | +0.012 |
Neg logit contributions | |
bons | -0.014 |
growth | -0.012 |
talk | -0.012 |
bars | -0.012 |
ests | -0.011 |
Token | there |
Token ID | 612 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
pleased | +0.237 |
glad | +0.224 |
sorry | +0.221 |
willing | +0.216 |
ok | +0.216 |
Neg logit contributions | |
boxes | -0.164 |
board | -0.159 |
els | -0.156 |
el | -0.155 |
blocks | -0.155 |
Token | was |
Token ID | 373 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
glad | +0.119 |
pleased | +0.117 |
impressed | +0.111 |
sorry | +0.111 |
okay | +0.105 |
Neg logit contributions | |
bons | -0.058 |
growth | -0.051 |
bars | -0.051 |
talk | -0.048 |
ests | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
glad | +0.035 |
pleased | +0.035 |
impressed | +0.032 |
sorry | +0.030 |
lucky | +0.030 |
Neg logit contributions | |
bons | -0.033 |
growth | -0.029 |
bars | -0.028 |
talk | -0.027 |
shirts | -0.026 |
Token | little |
Token ID | 1310 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
glad | +0.100 |
pleased | +0.099 |
impressed | +0.089 |
okay | +0.088 |
sorry | +0.086 |
Neg logit contributions | |
bons | -0.107 |
bars | -0.105 |
shirts | -0.101 |
growth | -0.100 |
board | -0.098 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
glad | +0.013 |
pleased | +0.012 |
impressed | +0.011 |
okay | +0.011 |
sorry | +0.011 |
Neg logit contributions | |
bons | -0.010 |
talk | -0.009 |
growth | -0.009 |
bars | -0.009 |
shirts | -0.008 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
glad | +0.123 |
pleased | +0.116 |
impressed | +0.110 |
okay | +0.106 |
sorry | +0.101 |
Neg logit contributions | |
bons | -0.128 |
bars | -0.116 |
growth | -0.116 |
talk | -0.111 |
board | -0.108 |
Token | day |
Token ID | 1110 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
glad | +0.097 |
pleased | +0.090 |
impressed | +0.087 |
okay | +0.084 |
sorry | +0.083 |
Neg logit contributions | |
bons | -0.077 |
growth | -0.065 |
talk | -0.064 |
bars | -0.063 |
shirts | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
glad | +0.108 |
pleased | +0.102 |
impressed | +0.096 |
sorry | +0.092 |
okay | +0.090 |
Neg logit contributions | |
bons | -0.109 |
talk | -0.093 |
bars | -0.093 |
growth | -0.091 |
als | -0.090 |
Token | I |
Token ID | 314 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
glad | +0.118 |
pleased | +0.113 |
impressed | +0.102 |
okay | +0.101 |
sorry | +0.097 |
Neg logit contributions | |
bons | -0.131 |
bars | -0.111 |
growth | -0.110 |
als | -0.109 |
talk | -0.106 |
Token | 'm |
Token ID | 1101 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
glad | +0.095 |
pleased | +0.089 |
impressed | +0.085 |
sorry | +0.082 |
okay | +0.080 |
Neg logit contributions | |
bons | -0.081 |
talk | -0.077 |
growth | -0.073 |
bars | -0.070 |
als | -0.067 |
Token | very |
Token ID | 845 |
Feature activation | +5.08e+00 |
Pos logit contributions | |
pleased | +0.143 |
shadow | +0.135 |
reetings | +0.132 |
impressed | +0.127 |
okay | +0.125 |
Neg logit contributions | |
bons | -0.205 |
stick | -0.182 |
al | -0.181 |
bars | -0.177 |
growth | -0.176 |
Token | enthusiastic |
Token ID | 21848 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
shadow | +0.119 |
okay | +0.118 |
reetings | +0.112 |
ok | +0.102 |
elcome | +0.098 |
Neg logit contributions | |
waters | -0.214 |
water | -0.207 |
leaves | -0.204 |
bons | -0.204 |
paper | -0.202 |
Token | about |
Token ID | 546 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
glad | +0.139 |
pleased | +0.132 |
impressed | +0.122 |
lucky | +0.119 |
sorry | +0.118 |
Neg logit contributions | |
bons | -0.115 |
talk | -0.095 |
bars | -0.093 |
pages | -0.089 |
shirts | -0.087 |
Token | n |
Token ID | 299 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
pleased | +0.151 |
glad | +0.150 |
impressed | +0.133 |
okay | +0.130 |
sorry | +0.128 |
Neg logit contributions | |
bons | -0.152 |
growth | -0.127 |
bars | -0.124 |
shirts | -0.118 |
pages | -0.116 |
Token | apping |
Token ID | 5912 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
glad | +0.067 |
pleased | +0.064 |
lucky | +0.059 |
sorry | +0.057 |
okay | +0.056 |
Neg logit contributions | |
bons | -0.091 |
talk | -0.086 |
growth | -0.083 |
ests | -0.081 |
room | -0.078 |
Token | in |
Token ID | 287 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
glad | +0.073 |
pleased | +0.068 |
impressed | +0.063 |
lucky | +0.061 |
sorry | +0.061 |
Neg logit contributions | |
bons | -0.074 |
talk | -0.063 |
growth | -0.061 |
bars | -0.061 |
shirts | -0.060 |
Token | 'm |
Token ID | 1101 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
glad | +0.063 |
pleased | +0.059 |
impressed | +0.056 |
sorry | +0.056 |
okay | +0.054 |
Neg logit contributions | |
bons | -0.053 |
talk | -0.049 |
growth | -0.047 |
bars | -0.046 |
als | -0.043 |
Token | flying |
Token ID | 7348 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
pleased | +0.165 |
reetings | +0.155 |
va | +0.153 |
concerned | +0.153 |
mber | +0.153 |
Neg logit contributions | |
bons | -0.153 |
bars | -0.126 |
leaves | -0.123 |
air | -0.113 |
aled | -0.113 |
Token | ! |
Token ID | 0 |
Feature activation | +9.27e-01 |
Pos logit contributions | |
glad | +0.088 |
pleased | +0.081 |
sorry | +0.074 |
impressed | +0.073 |
lucky | +0.072 |
Neg logit contributions | |
bons | -0.092 |
talk | -0.080 |
bars | -0.079 |
growth | -0.076 |
als | -0.076 |
Token | I |
Token ID | 314 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
glad | +0.050 |
pleased | +0.049 |
impressed | +0.045 |
lucky | +0.044 |
sorry | +0.042 |
Neg logit contributions | |
bons | -0.052 |
growth | -0.046 |
talk | -0.046 |
bars | -0.044 |
als | -0.043 |
Token | 'm |
Token ID | 1101 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
glad | +0.060 |
pleased | +0.057 |
impressed | +0.053 |
sorry | +0.053 |
okay | +0.051 |
Neg logit contributions | |
bons | -0.057 |
talk | -0.052 |
growth | -0.051 |
bars | -0.050 |
als | -0.047 |
Token | really |
Token ID | 1107 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
pleased | +0.178 |
reetings | +0.171 |
va | +0.166 |
mber | +0.165 |
glad | +0.162 |
Neg logit contributions | |
bons | -0.158 |
leaves | -0.125 |
bars | -0.123 |
air | -0.113 |
aled | -0.112 |
Token | flying |
Token ID | 7348 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
okay | +0.156 |
ok | +0.148 |
importantly | +0.138 |
glad | +0.136 |
ama | +0.135 |
Neg logit contributions | |
bons | -0.233 |
leaves | -0.207 |
stones | -0.201 |
air | -0.197 |
pictures | -0.194 |
Token | !" |
Token ID | 2474 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
glad | +0.078 |
pleased | +0.072 |
sorry | +0.065 |
impressed | +0.064 |
lucky | +0.063 |
Neg logit contributions | |
bons | -0.084 |
talk | -0.073 |
bars | -0.072 |
growth | -0.070 |
als | -0.068 |
Token | shouted |
Token ID | 17293 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
bons | +0.058 |
talk | +0.052 |
growth | +0.051 |
ests | +0.050 |
bars | +0.049 |
Neg logit contributions | |
glad | -0.050 |
pleased | -0.046 |
impressed | -0.042 |
sorry | -0.042 |
lucky | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +5.35e-01 |
Pos logit contributions | |
glad | +0.063 |
pleased | +0.058 |
impressed | +0.054 |
okay | +0.053 |
lucky | +0.053 |
Neg logit contributions | |
bons | -0.046 |
growth | -0.042 |
talk | -0.042 |
bars | -0.039 |
als | -0.038 |
Token | little |
Token ID | 1310 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
glad | +0.031 |
pleased | +0.029 |
impressed | +0.026 |
okay | +0.026 |
sorry | +0.026 |
Neg logit contributions | |
bons | -0.029 |
growth | -0.026 |
bars | -0.025 |
talk | -0.024 |
als | -0.024 |
Token | little |
Token ID | 1310 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
bons | +0.005 |
talk | +0.004 |
growth | +0.004 |
bars | +0.004 |
shirts | +0.004 |
Neg logit contributions | |
glad | -0.007 |
pleased | -0.006 |
impressed | -0.006 |
okay | -0.006 |
lucky | -0.006 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
bons | +0.012 |
talk | +0.010 |
growth | +0.010 |
bars | +0.010 |
shirts | +0.009 |
Neg logit contributions | |
glad | -0.014 |
pleased | -0.013 |
impressed | -0.012 |
okay | -0.012 |
lucky | -0.012 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
glad | +0.060 |
pleased | +0.056 |
okay | +0.052 |
impressed | +0.052 |
sorry | +0.051 |
Neg logit contributions | |
bons | -0.053 |
bars | -0.046 |
talk | -0.046 |
growth | -0.044 |
shirts | -0.043 |
Token | good |
Token ID | 922 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
glad | +0.070 |
pleased | +0.066 |
impressed | +0.061 |
shadow | +0.059 |
okay | +0.056 |
Neg logit contributions | |
bons | -0.128 |
growth | -0.108 |
bars | -0.107 |
fabric | -0.106 |
cans | -0.106 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.72e-02 |
Pos logit contributions | |
glad | +0.006 |
pleased | +0.006 |
impressed | +0.005 |
lucky | +0.005 |
sorry | +0.005 |
Neg logit contributions | |
bons | -0.005 |
talk | -0.005 |
growth | -0.004 |
bars | -0.004 |
shirts | -0.004 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
bons | +0.006 |
talk | +0.005 |
growth | +0.005 |
bars | +0.005 |
ests | +0.005 |
Neg logit contributions | |
glad | -0.005 |
pleased | -0.005 |
impressed | -0.005 |
okay | -0.004 |
lucky | -0.004 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
pleased | +0.213 |
glad | +0.201 |
ok | +0.197 |
sorry | +0.195 |
okay | +0.194 |
Neg logit contributions | |
boxes | -0.173 |
els | -0.171 |
talk | -0.169 |
board | -0.168 |
growth | -0.167 |
Token | a |
Token ID | 257 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
glad | +0.040 |
pleased | +0.038 |
impressed | +0.035 |
sorry | +0.034 |
okay | +0.033 |
Neg logit contributions | |
bons | -0.035 |
growth | -0.031 |
talk | -0.030 |
bars | -0.030 |
ests | -0.028 |
Token | time |
Token ID | 640 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
glad | +0.071 |
pleased | +0.069 |
impressed | +0.064 |
sorry | +0.062 |
lucky | +0.059 |
Neg logit contributions | |
bons | -0.073 |
growth | -0.065 |
bars | -0.063 |
talk | -0.063 |
shirts | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
pleased | +0.238 |
impressed | +0.218 |
glad | +0.215 |
blessed | +0.202 |
thankful | +0.201 |
Neg logit contributions | |
room | -0.094 |
bons | -0.084 |
bars | -0.082 |
als | -0.071 |
ests | -0.070 |
Token | there |
Token ID | 612 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
pleased | +0.155 |
glad | +0.142 |
impressed | +0.135 |
sorry | +0.125 |
blessed | +0.119 |
Neg logit contributions | |
bons | -0.134 |
bars | -0.132 |
x | -0.130 |
al | -0.128 |
room | -0.128 |
Token | together |
Token ID | 1978 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
bons | +0.024 |
talk | +0.021 |
growth | +0.021 |
bars | +0.020 |
ests | +0.020 |
Neg logit contributions | |
glad | -0.018 |
pleased | -0.017 |
impressed | -0.015 |
okay | -0.015 |
lucky | -0.015 |
Token | all |
Token ID | 477 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
bons | +0.028 |
talk | +0.025 |
growth | +0.025 |
bars | +0.024 |
ests | +0.023 |
Neg logit contributions | |
glad | -0.026 |
pleased | -0.025 |
impressed | -0.023 |
okay | -0.022 |
lucky | -0.022 |
Token | afternoon |
Token ID | 6672 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
glad | +0.002 |
pleased | +0.002 |
okay | +0.002 |
sorry | +0.002 |
impressed | +0.002 |
Neg logit contributions | |
bons | -0.003 |
talk | -0.003 |
growth | -0.002 |
bars | -0.002 |
ales | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
glad | +0.051 |
pleased | +0.048 |
okay | +0.045 |
sorry | +0.044 |
lucky | +0.044 |
Neg logit contributions | |
bons | -0.043 |
bars | -0.037 |
talk | -0.037 |
growth | -0.036 |
shirts | -0.035 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.70e-03 |
Pos logit contributions | |
glad | +0.064 |
pleased | +0.062 |
impressed | +0.059 |
sorry | +0.057 |
lucky | +0.056 |
Neg logit contributions | |
bons | -0.047 |
growth | -0.043 |
talk | -0.039 |
bars | -0.039 |
als | -0.038 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
bons | +0.000 |
talk | +0.000 |
growth | +0.000 |
bars | +0.000 |
ests | +0.000 |
Neg logit contributions | |
glad | -0.000 |
pleased | -0.000 |
impressed | -0.000 |
okay | -0.000 |
sorry | -0.000 |
Token | upon |
Token ID | 2402 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
pleased | +0.213 |
glad | +0.205 |
impressed | +0.194 |
sorry | +0.194 |
shadow | +0.185 |
Neg logit contributions | |
els | -0.186 |
ales | -0.183 |
als | -0.183 |
berries | -0.182 |
ens | -0.181 |
Token | a |
Token ID | 257 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
glad | +0.045 |
pleased | +0.043 |
impressed | +0.040 |
sorry | +0.039 |
lucky | +0.037 |
Neg logit contributions | |
bons | -0.037 |
growth | -0.034 |
bars | -0.032 |
talk | -0.031 |
shirts | -0.029 |
Token | time |
Token ID | 640 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
glad | +0.089 |
pleased | +0.087 |
impressed | +0.081 |
sorry | +0.080 |
lucky | +0.075 |
Neg logit contributions | |
bons | -0.084 |
growth | -0.075 |
talk | -0.072 |
bars | -0.072 |
shirts | -0.070 |
Token | there |
Token ID | 612 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
pleased | +0.256 |
glad | +0.242 |
impressed | +0.235 |
thankful | +0.222 |
grateful | +0.219 |
Neg logit contributions | |
room | -0.088 |
bons | -0.085 |
bars | -0.085 |
cans | -0.080 |
growth | -0.079 |
Token | was |
Token ID | 373 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
glad | +0.050 |
pleased | +0.048 |
sorry | +0.044 |
impressed | +0.044 |
okay | +0.043 |
Neg logit contributions | |
bons | -0.032 |
growth | -0.028 |
bars | -0.027 |
talk | -0.027 |
ests | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
glad | +0.002 |
pleased | +0.001 |
okay | +0.001 |
sorry | +0.001 |
impressed | +0.001 |
Neg logit contributions | |
bons | -0.003 |
talk | -0.003 |
growth | -0.003 |
bars | -0.003 |
ests | -0.003 |
Token | never |
Token ID | 1239 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
glad | +0.013 |
pleased | +0.012 |
impressed | +0.011 |
lucky | +0.011 |
okay | +0.011 |
Neg logit contributions | |
bons | -0.010 |
talk | -0.008 |
growth | -0.008 |
bars | -0.008 |
shirts | -0.008 |
Token | trust |
Token ID | 3774 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
glad | +0.100 |
pleased | +0.095 |
impressed | +0.088 |
lucky | +0.087 |
sorry | +0.084 |
Neg logit contributions | |
bons | -0.065 |
bars | -0.057 |
talk | -0.057 |
growth | -0.055 |
ales | -0.050 |
Token | someone |
Token ID | 2130 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
glad | +0.031 |
pleased | +0.030 |
impressed | +0.028 |
lucky | +0.027 |
sorry | +0.027 |
Neg logit contributions | |
bons | -0.025 |
growth | -0.021 |
talk | -0.021 |
bars | -0.020 |
ales | -0.020 |
Token | who |
Token ID | 508 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
glad | +0.134 |
pleased | +0.128 |
impressed | +0.120 |
sorry | +0.117 |
lucky | +0.113 |
Neg logit contributions | |
bons | -0.094 |
bars | -0.085 |
talk | -0.081 |
als | -0.077 |
growth | -0.075 |
Token | seems |
Token ID | 2331 |
Feature activation | +4.79e+00 |
Pos logit contributions | |
glad | +0.073 |
pleased | +0.070 |
impressed | +0.065 |
okay | +0.064 |
sorry | +0.063 |
Neg logit contributions | |
bons | -0.065 |
talk | -0.058 |
bars | -0.057 |
growth | -0.056 |
als | -0.052 |
Token | too |
Token ID | 1165 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
glad | +0.241 |
pleased | +0.226 |
shadow | +0.208 |
sorry | +0.201 |
okay | +0.201 |
Neg logit contributions | |
bons | -0.188 |
bars | -0.170 |
talk | -0.168 |
cans | -0.167 |
targets | -0.166 |
Token | good |
Token ID | 922 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
glad | +0.162 |
pleased | +0.150 |
sorry | +0.135 |
impressed | +0.132 |
okay | +0.131 |
Neg logit contributions | |
bons | -0.168 |
bars | -0.154 |
cans | -0.153 |
growth | -0.152 |
pages | -0.152 |
Token | to |
Token ID | 284 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
glad | +0.053 |
pleased | +0.050 |
impressed | +0.046 |
lucky | +0.045 |
sorry | +0.044 |
Neg logit contributions | |
bons | -0.039 |
talk | -0.035 |
bars | -0.032 |
growth | -0.032 |
shirts | -0.031 |
Token | be |
Token ID | 307 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
glad | +0.073 |
pleased | +0.071 |
impressed | +0.065 |
lucky | +0.063 |
sorry | +0.061 |
Neg logit contributions | |
bons | -0.053 |
talk | -0.046 |
growth | -0.045 |
bars | -0.044 |
ales | -0.041 |
Token | true |
Token ID | 2081 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
glad | +0.102 |
pleased | +0.094 |
impressed | +0.086 |
lucky | +0.083 |
sorry | +0.082 |
Neg logit contributions | |
bons | -0.110 |
talk | -0.093 |
bars | -0.092 |
growth | -0.088 |
cans | -0.085 |
Token | often |
Token ID | 1690 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
bons | +0.068 |
growth | +0.060 |
talk | +0.059 |
ests | +0.057 |
bars | +0.055 |
Neg logit contributions | |
glad | -0.072 |
pleased | -0.067 |
impressed | -0.062 |
okay | -0.061 |
sorry | -0.061 |
Token | spent |
Token ID | 3377 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
bons | +0.035 |
talk | +0.031 |
growth | +0.031 |
bars | +0.029 |
ests | +0.028 |
Neg logit contributions | |
glad | -0.045 |
pleased | -0.042 |
okay | -0.039 |
impressed | -0.039 |
lucky | -0.039 |
Token | time |
Token ID | 640 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
bons | +0.043 |
talk | +0.039 |
growth | +0.038 |
bars | +0.036 |
ests | +0.036 |
Neg logit contributions | |
glad | -0.041 |
pleased | -0.038 |
lucky | -0.034 |
impressed | -0.034 |
okay | -0.034 |
Token | together |
Token ID | 1978 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
glad | +0.114 |
pleased | +0.109 |
lucky | +0.103 |
okay | +0.101 |
sorry | +0.100 |
Neg logit contributions | |
bons | -0.104 |
bars | -0.093 |
talk | -0.092 |
shirts | -0.088 |
cans | -0.081 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
glad | +0.014 |
pleased | +0.013 |
impressed | +0.012 |
sorry | +0.012 |
lucky | +0.012 |
Neg logit contributions | |
bons | -0.014 |
talk | -0.013 |
growth | -0.013 |
bars | -0.013 |
shirts | -0.012 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
bons | +0.029 |
talk | +0.026 |
growth | +0.026 |
bars | +0.025 |
ales | +0.024 |
Neg logit contributions | |
glad | -0.024 |
pleased | -0.022 |
okay | -0.020 |
impressed | -0.020 |
lucky | -0.020 |
Token | there |
Token ID | 612 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
pleased | +0.213 |
glad | +0.201 |
sorry | +0.197 |
ok | +0.195 |
okay | +0.195 |
Neg logit contributions | |
boxes | -0.182 |
els | -0.181 |
x | -0.177 |
oline | -0.175 |
talk | -0.174 |
Token | was |
Token ID | 373 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
glad | +0.127 |
pleased | +0.125 |
sorry | +0.118 |
impressed | +0.118 |
lucky | +0.111 |
Neg logit contributions | |
bons | -0.063 |
bars | -0.056 |
growth | -0.054 |
ests | -0.051 |
talk | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
bons | +0.001 |
growth | +0.001 |
talk | +0.001 |
bars | +0.001 |
shirts | +0.001 |
Neg logit contributions | |
glad | -0.001 |
pleased | -0.001 |
impressed | -0.001 |
sorry | -0.001 |
okay | -0.001 |
Token | little |
Token ID | 1310 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
glad | +0.102 |
pleased | +0.101 |
okay | +0.090 |
impressed | +0.089 |
ok | +0.089 |
Neg logit contributions | |
bons | -0.109 |
bars | -0.106 |
shirts | -0.103 |
growth | -0.101 |
cash | -0.098 |
Token | owl |
Token ID | 39610 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
glad | +0.013 |
pleased | +0.013 |
impressed | +0.012 |
okay | +0.012 |
sorry | +0.011 |
Neg logit contributions | |
bons | -0.011 |
talk | -0.009 |
growth | -0.009 |
bars | -0.009 |
shirts | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | +8.47e-02 |
Pos logit contributions | |
glad | +0.029 |
pleased | +0.027 |
impressed | +0.025 |
lucky | +0.025 |
okay | +0.025 |
Neg logit contributions | |
bons | -0.026 |
talk | -0.023 |
growth | -0.023 |
bars | -0.022 |
als | -0.022 |
Token | his |
Token ID | 465 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
glad | +0.005 |
pleased | +0.005 |
impressed | +0.005 |
okay | +0.005 |
lucky | +0.005 |
Neg logit contributions | |
bons | -0.004 |
growth | -0.004 |
talk | -0.004 |
bars | -0.003 |
shirts | -0.003 |
Token | face |
Token ID | 1986 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
glad | +0.010 |
pleased | +0.010 |
impressed | +0.009 |
okay | +0.009 |
lucky | +0.009 |
Neg logit contributions | |
bons | -0.008 |
growth | -0.007 |
talk | -0.007 |
bars | -0.006 |
shirts | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
bons | +0.012 |
talk | +0.010 |
growth | +0.010 |
bars | +0.010 |
shirts | +0.009 |
Neg logit contributions | |
glad | -0.013 |
pleased | -0.012 |
impressed | -0.011 |
okay | -0.011 |
lucky | -0.011 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
glad | +0.073 |
pleased | +0.072 |
impressed | +0.067 |
lucky | +0.066 |
okay | +0.065 |
Neg logit contributions | |
bons | -0.056 |
growth | -0.048 |
talk | -0.045 |
ests | -0.044 |
bars | -0.043 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
glad | +0.012 |
pleased | +0.011 |
impressed | +0.010 |
okay | +0.010 |
sorry | +0.010 |
Neg logit contributions | |
bons | -0.013 |
talk | -0.012 |
growth | -0.012 |
bars | -0.011 |
ests | -0.011 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
pleased | +0.219 |
glad | +0.208 |
sorry | +0.206 |
ok | +0.204 |
impressed | +0.200 |
Neg logit contributions | |
boxes | -0.166 |
ests | -0.165 |
el | -0.162 |
board | -0.162 |
els | -0.161 |
Token | a |
Token ID | 257 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
glad | +0.035 |
pleased | +0.034 |
impressed | +0.031 |
sorry | +0.030 |
okay | +0.030 |
Neg logit contributions | |
bons | -0.031 |
growth | -0.028 |
talk | -0.027 |
bars | -0.027 |
ests | -0.025 |
Token | time |
Token ID | 640 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
glad | +0.075 |
pleased | +0.073 |
impressed | +0.068 |
sorry | +0.066 |
lucky | +0.062 |
Neg logit contributions | |
bons | -0.074 |
growth | -0.066 |
talk | -0.064 |
bars | -0.064 |
shirts | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
pleased | +0.244 |
glad | +0.227 |
impressed | +0.224 |
thankful | +0.208 |
blessed | +0.207 |
Neg logit contributions | |
room | -0.097 |
bons | -0.092 |
bars | -0.087 |
ests | -0.079 |
cans | -0.079 |
Token | there |
Token ID | 612 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
pleased | +0.172 |
glad | +0.157 |
impressed | +0.151 |
sorry | +0.139 |
blessed | +0.136 |
Neg logit contributions | |
bons | -0.136 |
bars | -0.133 |
room | -0.131 |
x | -0.131 |
els | -0.129 |
Token | |
Token ID | 220 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
bons | +0.124 |
growth | +0.110 |
talk | +0.108 |
stations | +0.103 |
ales | +0.103 |
Neg logit contributions | |
glad | -0.104 |
pleased | -0.095 |
sorry | -0.089 |
impressed | -0.085 |
okay | -0.084 |
Token |
|
Token ID | 198 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
glad | +0.083 |
pleased | +0.081 |
impressed | +0.073 |
lucky | +0.072 |
sorry | +0.071 |
Neg logit contributions | |
bons | -0.069 |
talk | -0.062 |
bars | -0.061 |
growth | -0.060 |
ests | -0.058 |
Token | The |
Token ID | 464 |
Feature activation | +5.17e-04 |
Pos logit contributions | |
glad | +0.019 |
pleased | +0.017 |
sorry | +0.016 |
okay | +0.016 |
lucky | +0.016 |
Neg logit contributions | |
bons | -0.020 |
talk | -0.018 |
growth | -0.017 |
bars | -0.017 |
ests | -0.016 |
Token | boy |
Token ID | 2933 |
Feature activation | -8.44e-02 |
Pos logit contributions | |
glad | +0.000 |
pleased | +0.000 |
impressed | +0.000 |
okay | +0.000 |
sorry | +0.000 |
Neg logit contributions | |
bons | -0.000 |
talk | -0.000 |
growth | -0.000 |
bars | -0.000 |
shirts | -0.000 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
bons | +0.004 |
talk | +0.004 |
growth | +0.004 |
bars | +0.004 |
shirts | +0.004 |
Neg logit contributions | |
glad | -0.005 |
pleased | -0.005 |
okay | -0.004 |
impressed | -0.004 |
lucky | -0.004 |
Token | : |
Token ID | 25 |
Feature activation | +4.75e+00 |
Pos logit contributions | |
glad | +0.013 |
pleased | +0.012 |
impressed | +0.011 |
lucky | +0.011 |
okay | +0.010 |
Neg logit contributions | |
bons | -0.014 |
talk | -0.013 |
growth | -0.013 |
bars | -0.012 |
ests | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | -8.90e-01 |
Pos logit contributions | |
pleased | +0.281 |
glad | +0.281 |
lucky | +0.270 |
blessed | +0.268 |
alright | +0.261 |
Neg logit contributions | |
bons | -0.108 |
als | -0.107 |
shirts | -0.106 |
cans | -0.104 |
board | -0.104 |
Token | What |
Token ID | 2061 |
Feature activation | +6.39e-01 |
Pos logit contributions | |
bons | +0.049 |
growth | +0.042 |
talk | +0.042 |
bars | +0.042 |
ales | +0.040 |
Neg logit contributions | |
glad | -0.051 |
pleased | -0.049 |
impressed | -0.045 |
okay | -0.045 |
lucky | -0.044 |
Token | do |
Token ID | 466 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
glad | +0.042 |
pleased | +0.040 |
impressed | +0.036 |
sorry | +0.035 |
lucky | +0.035 |
Neg logit contributions | |
bons | -0.029 |
bars | -0.025 |
talk | -0.025 |
growth | -0.024 |
shirts | -0.024 |
Token | you |
Token ID | 345 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
glad | +0.092 |
pleased | +0.087 |
impressed | +0.080 |
sorry | +0.073 |
okay | +0.073 |
Neg logit contributions | |
bons | -0.087 |
talk | -0.075 |
bars | -0.074 |
growth | -0.074 |
ests | -0.071 |
Token | mean |
Token ID | 1612 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
glad | +0.038 |
pleased | +0.037 |
impressed | +0.033 |
lucky | +0.032 |
sorry | +0.032 |
Neg logit contributions | |
bons | -0.034 |
talk | -0.029 |
growth | -0.029 |
bars | -0.028 |
ales | -0.027 |
Token | up |
Token ID | 510 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
bons | +0.039 |
growth | +0.035 |
talk | +0.034 |
bars | +0.033 |
ests | +0.032 |
Neg logit contributions | |
glad | -0.035 |
pleased | -0.033 |
impressed | -0.030 |
lucky | -0.029 |
okay | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
bons | +0.115 |
growth | +0.101 |
ests | +0.101 |
talk | +0.100 |
ales | +0.098 |
Neg logit contributions | |
glad | -0.059 |
pleased | -0.054 |
impressed | -0.048 |
lucky | -0.045 |
happy | -0.043 |
Token | him |
Token ID | 683 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
bons | +0.042 |
talk | +0.037 |
growth | +0.036 |
bars | +0.035 |
ests | +0.035 |
Neg logit contributions | |
glad | -0.045 |
pleased | -0.042 |
lucky | -0.038 |
okay | -0.038 |
sorry | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
bons | +0.012 |
talk | +0.010 |
growth | +0.010 |
bars | +0.010 |
shirts | +0.009 |
Neg logit contributions | |
glad | -0.017 |
pleased | -0.017 |
impressed | -0.015 |
okay | -0.015 |
lucky | -0.015 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
glad | +0.082 |
pleased | +0.077 |
okay | +0.075 |
sorry | +0.074 |
lucky | +0.072 |
Neg logit contributions | |
bons | -0.048 |
talk | -0.046 |
bars | -0.042 |
ted | -0.042 |
als | -0.042 |
Token | : |
Token ID | 25 |
Feature activation | +4.75e+00 |
Pos logit contributions | |
glad | +0.056 |
pleased | +0.054 |
impressed | +0.049 |
lucky | +0.047 |
sorry | +0.045 |
Neg logit contributions | |
bons | -0.053 |
talk | -0.048 |
bars | -0.048 |
cans | -0.046 |
shirts | -0.046 |
Token | " |
Token ID | 366 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
pleased | +0.262 |
glad | +0.260 |
lucky | +0.248 |
blessed | +0.244 |
impressed | +0.235 |
Neg logit contributions | |
bons | -0.136 |
cans | -0.130 |
growth | -0.130 |
als | -0.126 |
shirts | -0.124 |
Token | Why |
Token ID | 5195 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
bons | +0.046 |
talk | +0.039 |
growth | +0.039 |
bars | +0.038 |
ests | +0.037 |
Neg logit contributions | |
glad | -0.048 |
pleased | -0.046 |
okay | -0.042 |
impressed | -0.042 |
lucky | -0.041 |
Token | are |
Token ID | 389 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
glad | +0.073 |
pleased | +0.071 |
impressed | +0.066 |
lucky | +0.063 |
sorry | +0.063 |
Neg logit contributions | |
bons | -0.027 |
bars | -0.022 |
talk | -0.021 |
shirts | -0.020 |
ests | -0.020 |
Token | you |
Token ID | 345 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
glad | +0.071 |
pleased | +0.069 |
impressed | +0.061 |
okay | +0.058 |
lucky | +0.057 |
Neg logit contributions | |
bons | -0.068 |
shirts | -0.061 |
cans | -0.061 |
bars | -0.061 |
talk | -0.061 |
Token | cl |
Token ID | 537 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
pleased | +0.109 |
glad | +0.103 |
blessed | +0.089 |
impressed | +0.088 |
lucky | +0.088 |
Neg logit contributions | |
bons | -0.110 |
stick | -0.106 |
bars | -0.103 |
shirts | -0.102 |
pedals | -0.100 |
Token | lonely |
Token ID | 21757 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
glad | +0.127 |
pleased | +0.125 |
impressed | +0.116 |
shadow | +0.113 |
okay | +0.109 |
Neg logit contributions | |
bons | -0.145 |
bars | -0.141 |
growth | -0.135 |
talk | -0.133 |
stations | -0.128 |
Token | cloud |
Token ID | 6279 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
glad | +0.114 |
pleased | +0.108 |
impressed | +0.102 |
okay | +0.096 |
lucky | +0.092 |
Neg logit contributions | |
bons | -0.081 |
bars | -0.074 |
talk | -0.072 |
growth | -0.072 |
board | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
glad | +0.123 |
pleased | +0.116 |
impressed | +0.108 |
lucky | +0.106 |
sorry | +0.102 |
Neg logit contributions | |
bons | -0.113 |
talk | -0.101 |
bars | -0.101 |
als | -0.098 |
board | -0.093 |
Token | I |
Token ID | 314 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
glad | +0.136 |
pleased | +0.135 |
impressed | +0.121 |
okay | +0.116 |
lucky | +0.112 |
Neg logit contributions | |
bons | -0.138 |
bars | -0.122 |
als | -0.120 |
growth | -0.117 |
talk | -0.115 |
Token | 'm |
Token ID | 1101 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
glad | +0.077 |
pleased | +0.073 |
impressed | +0.068 |
sorry | +0.065 |
lucky | +0.064 |
Neg logit contributions | |
bons | -0.072 |
talk | -0.069 |
bars | -0.064 |
growth | -0.063 |
als | -0.061 |
Token | here |
Token ID | 994 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
pleased | +0.159 |
reetings | +0.148 |
shadow | +0.139 |
impressed | +0.134 |
mber | +0.133 |
Neg logit contributions | |
bons | -0.189 |
stick | -0.180 |
bars | -0.175 |
al | -0.158 |
each | -0.158 |
Token | to |
Token ID | 284 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
glad | +0.292 |
pleased | +0.287 |
impressed | +0.278 |
sorry | +0.265 |
lucky | +0.258 |
Neg logit contributions | |
bons | -0.143 |
bars | -0.109 |
talk | -0.102 |
als | -0.101 |
each | -0.097 |
Token | visit |
Token ID | 3187 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
glad | +0.074 |
pleased | +0.072 |
impressed | +0.066 |
lucky | +0.064 |
okay | +0.062 |
Neg logit contributions | |
bons | -0.048 |
growth | -0.041 |
talk | -0.041 |
bars | -0.038 |
stick | -0.034 |
Token | you |
Token ID | 345 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
glad | +0.071 |
pleased | +0.068 |
impressed | +0.062 |
sorry | +0.059 |
lucky | +0.059 |
Neg logit contributions | |
bons | -0.065 |
talk | -0.056 |
bars | -0.054 |
growth | -0.054 |
ales | -0.052 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
glad | +0.116 |
pleased | +0.109 |
impressed | +0.100 |
sorry | +0.097 |
lucky | +0.096 |
Neg logit contributions | |
bons | -0.092 |
bars | -0.080 |
talk | -0.080 |
als | -0.077 |
growth | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
bons | +0.107 |
talk | +0.096 |
growth | +0.095 |
ales | +0.094 |
shirts | +0.092 |
Neg logit contributions | |
glad | -0.076 |
pleased | -0.068 |
sorry | -0.062 |
impressed | -0.059 |
lucky | -0.059 |
Token | wonderful |
Token ID | 7932 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
glad | +0.133 |
pleased | +0.124 |
impressed | +0.117 |
okay | +0.111 |
sorry | +0.106 |
Neg logit contributions | |
bons | -0.128 |
bars | -0.121 |
shirts | -0.112 |
pages | -0.110 |
board | -0.110 |
Token | boat |
Token ID | 8848 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
glad | +0.093 |
pleased | +0.087 |
impressed | +0.082 |
okay | +0.082 |
sorry | +0.079 |
Neg logit contributions | |
bons | -0.072 |
growth | -0.060 |
bars | -0.059 |
talk | -0.059 |
shirts | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
glad | +0.047 |
pleased | +0.044 |
impressed | +0.041 |
okay | +0.040 |
lucky | +0.040 |
Neg logit contributions | |
bons | -0.040 |
talk | -0.036 |
bars | -0.034 |
growth | -0.034 |
ales | -0.032 |
Token | I |
Token ID | 314 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
glad | +0.014 |
pleased | +0.013 |
impressed | +0.012 |
lucky | +0.012 |
okay | +0.012 |
Neg logit contributions | |
bons | -0.011 |
talk | -0.009 |
growth | -0.009 |
bars | -0.009 |
ests | -0.009 |
Token | 'm |
Token ID | 1101 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
glad | +0.096 |
pleased | +0.089 |
impressed | +0.085 |
sorry | +0.083 |
okay | +0.082 |
Neg logit contributions | |
bons | -0.083 |
talk | -0.081 |
bars | -0.072 |
growth | -0.072 |
ests | -0.067 |
Token | very |
Token ID | 845 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
pleased | +0.124 |
reetings | +0.116 |
impressed | +0.110 |
okay | +0.107 |
shadow | +0.106 |
Neg logit contributions | |
bons | -0.199 |
bars | -0.174 |
aled | -0.167 |
fabric | -0.161 |
stick | -0.160 |
Token | proud |
Token ID | 6613 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
reetings | +0.121 |
okay | +0.120 |
ok | +0.109 |
shadow | +0.108 |
alright | +0.100 |
Neg logit contributions | |
bons | -0.192 |
cans | -0.190 |
water | -0.189 |
bars | -0.189 |
waters | -0.188 |
Token | of |
Token ID | 286 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
glad | +0.079 |
pleased | +0.075 |
sorry | +0.070 |
okay | +0.070 |
impressed | +0.069 |
Neg logit contributions | |
bons | -0.055 |
bars | -0.045 |
talk | -0.045 |
shirts | -0.043 |
cans | -0.042 |
Token | you |
Token ID | 345 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
pleased | +0.068 |
glad | +0.068 |
impressed | +0.062 |
okay | +0.059 |
lucky | +0.057 |
Neg logit contributions | |
bons | -0.069 |
growth | -0.055 |
bars | -0.055 |
shirts | -0.052 |
pages | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
glad | +0.103 |
pleased | +0.101 |
impressed | +0.095 |
sorry | +0.089 |
lucky | +0.087 |
Neg logit contributions | |
bons | -0.087 |
bars | -0.076 |
talk | -0.071 |
shirts | -0.071 |
als | -0.070 |
Token | Let |
Token ID | 3914 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
glad | +0.024 |
pleased | +0.023 |
impressed | +0.021 |
lucky | +0.021 |
okay | +0.021 |
Neg logit contributions | |
bons | -0.026 |
growth | -0.023 |
talk | -0.022 |
bars | -0.021 |
ests | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
glad | +0.059 |
pleased | +0.055 |
impressed | +0.051 |
okay | +0.051 |
lucky | +0.050 |
Neg logit contributions | |
bons | -0.046 |
talk | -0.040 |
bars | -0.040 |
growth | -0.039 |
shirts | -0.037 |
Token | each |
Token ID | 1123 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
glad | +0.020 |
pleased | +0.019 |
okay | +0.017 |
impressed | +0.017 |
lucky | +0.017 |
Neg logit contributions | |
bons | -0.020 |
talk | -0.017 |
growth | -0.017 |
bars | -0.016 |
shirts | -0.016 |
Token | other |
Token ID | 584 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
bons | +0.007 |
talk | +0.006 |
growth | +0.006 |
bars | +0.006 |
ests | +0.006 |
Neg logit contributions | |
glad | -0.007 |
pleased | -0.007 |
impressed | -0.006 |
okay | -0.006 |
sorry | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
glad | +0.017 |
pleased | +0.016 |
impressed | +0.015 |
okay | +0.014 |
sorry | +0.014 |
Neg logit contributions | |
bons | -0.015 |
talk | -0.013 |
growth | -0.013 |
bars | -0.013 |
shirts | -0.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.32e-02 |
Pos logit contributions | |
glad | +0.009 |
pleased | +0.009 |
impressed | +0.008 |
lucky | +0.008 |
okay | +0.008 |
Neg logit contributions | |
bons | -0.008 |
growth | -0.007 |
talk | -0.007 |
bars | -0.006 |
ests | -0.006 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
glad | +0.005 |
pleased | +0.005 |
impressed | +0.004 |
okay | +0.004 |
sorry | +0.004 |
Neg logit contributions | |
bons | -0.006 |
talk | -0.005 |
growth | -0.005 |
bars | -0.005 |
ests | -0.005 |
Token | upon |
Token ID | 2402 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
pleased | +0.216 |
glad | +0.203 |
ok | +0.202 |
shadow | +0.197 |
okay | +0.196 |
Neg logit contributions | |
els | -0.169 |
boxes | -0.168 |
ests | -0.167 |
board | -0.166 |
blocks | -0.164 |
Token | a |
Token ID | 257 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
glad | +0.025 |
pleased | +0.024 |
impressed | +0.022 |
sorry | +0.021 |
okay | +0.021 |
Neg logit contributions | |
bons | -0.024 |
growth | -0.021 |
talk | -0.021 |
bars | -0.020 |
ests | -0.019 |
Token | time |
Token ID | 640 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
glad | +0.084 |
pleased | +0.082 |
impressed | +0.076 |
sorry | +0.074 |
lucky | +0.070 |
Neg logit contributions | |
bons | -0.080 |
growth | -0.071 |
bars | -0.069 |
talk | -0.069 |
shirts | -0.066 |
Token | , |
Token ID | 11 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
pleased | +0.247 |
glad | +0.230 |
impressed | +0.224 |
blessed | +0.211 |
thankful | +0.210 |
Neg logit contributions | |
room | -0.101 |
bons | -0.091 |
bars | -0.090 |
cans | -0.084 |
ests | -0.081 |
Token | a |
Token ID | 257 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
pleased | +0.175 |
glad | +0.161 |
impressed | +0.152 |
blessed | +0.141 |
sorry | +0.141 |
Neg logit contributions | |
bars | -0.122 |
bons | -0.120 |
room | -0.118 |
els | -0.118 |
x | -0.117 |
Token | help |
Token ID | 1037 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
bons | +0.005 |
talk | +0.004 |
growth | +0.004 |
bars | +0.004 |
shirts | +0.004 |
Neg logit contributions | |
glad | -0.005 |
pleased | -0.005 |
impressed | -0.005 |
lucky | -0.005 |
okay | -0.005 |
Token | out |
Token ID | 503 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
bons | +0.023 |
talk | +0.020 |
growth | +0.020 |
bars | +0.019 |
ests | +0.018 |
Neg logit contributions | |
glad | -0.024 |
pleased | -0.023 |
okay | -0.021 |
impressed | -0.021 |
lucky | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
glad | +0.084 |
pleased | +0.079 |
impressed | +0.073 |
sorry | +0.070 |
lucky | +0.070 |
Neg logit contributions | |
bons | -0.081 |
bars | -0.070 |
growth | -0.069 |
talk | -0.067 |
shirts | -0.066 |
Token | be |
Token ID | 307 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
glad | +0.083 |
pleased | +0.079 |
impressed | +0.073 |
lucky | +0.071 |
okay | +0.071 |
Neg logit contributions | |
bons | -0.060 |
talk | -0.052 |
bars | -0.051 |
growth | -0.050 |
shirts | -0.047 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
glad | +0.086 |
pleased | +0.077 |
sorry | +0.071 |
impressed | +0.068 |
lucky | +0.066 |
Neg logit contributions | |
bons | -0.161 |
bars | -0.130 |
talk | -0.130 |
boxes | -0.126 |
shirts | -0.125 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
bons | +0.012 |
talk | +0.010 |
growth | +0.010 |
bars | +0.010 |
ests | +0.010 |
Neg logit contributions | |
glad | -0.011 |
pleased | -0.010 |
impressed | -0.009 |
okay | -0.009 |
lucky | -0.009 |
Token | upon |
Token ID | 2402 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
pleased | +0.218 |
glad | +0.209 |
sorry | +0.202 |
impressed | +0.199 |
grateful | +0.192 |
Neg logit contributions | |
boxes | -0.165 |
board | -0.163 |
els | -0.162 |
ens | -0.159 |
talk | -0.156 |
Token | a |
Token ID | 257 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
glad | +0.057 |
pleased | +0.055 |
impressed | +0.051 |
sorry | +0.049 |
lucky | +0.048 |
Neg logit contributions | |
bons | -0.046 |
growth | -0.041 |
bars | -0.040 |
talk | -0.039 |
ests | -0.036 |
Token | time |
Token ID | 640 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
glad | +0.094 |
pleased | +0.091 |
impressed | +0.086 |
sorry | +0.083 |
lucky | +0.079 |
Neg logit contributions | |
bons | -0.089 |
growth | -0.078 |
bars | -0.077 |
talk | -0.076 |
shirts | -0.072 |
Token | there |
Token ID | 612 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
pleased | +0.279 |
impressed | +0.259 |
glad | +0.251 |
blessed | +0.244 |
thankful | +0.239 |
Neg logit contributions | |
room | -0.088 |
bars | -0.070 |
bons | -0.066 |
blocks | -0.059 |
benches | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
glad | +0.034 |
pleased | +0.033 |
impressed | +0.030 |
sorry | +0.030 |
okay | +0.030 |
Neg logit contributions | |
bons | -0.024 |
growth | -0.021 |
talk | -0.021 |
bars | -0.021 |
ests | -0.019 |
Token | tasted |
Token ID | 29187 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
glad | +0.054 |
pleased | +0.052 |
impressed | +0.048 |
okay | +0.045 |
lucky | +0.044 |
Neg logit contributions | |
bons | -0.074 |
bars | -0.064 |
talk | -0.064 |
cans | -0.061 |
shirts | -0.061 |
Token | y |
Token ID | 331 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
glad | +0.063 |
pleased | +0.059 |
impressed | +0.054 |
sorry | +0.052 |
lucky | +0.052 |
Neg logit contributions | |
bons | -0.061 |
bars | -0.052 |
shirts | -0.051 |
talk | -0.050 |
cans | -0.050 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
bons | +0.106 |
talk | +0.101 |
growth | +0.101 |
bars | +0.094 |
stations | +0.091 |
Neg logit contributions | |
glad | -0.068 |
pleased | -0.063 |
okay | -0.059 |
lucky | -0.056 |
sorry | -0.055 |
Token | ! |
Token ID | 0 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
bons | +0.011 |
talk | +0.010 |
growth | +0.010 |
bars | +0.009 |
shirts | +0.009 |
Neg logit contributions | |
glad | -0.015 |
pleased | -0.014 |
impressed | -0.013 |
okay | -0.013 |
lucky | -0.013 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
glad | +0.032 |
pleased | +0.030 |
impressed | +0.028 |
lucky | +0.028 |
okay | +0.027 |
Neg logit contributions | |
bons | -0.030 |
growth | -0.026 |
talk | -0.026 |
bars | -0.025 |
ests | -0.024 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
glad | +0.027 |
pleased | +0.026 |
impressed | +0.024 |
sorry | +0.023 |
okay | +0.023 |
Neg logit contributions | |
bons | -0.028 |
growth | -0.025 |
talk | -0.025 |
bars | -0.024 |
ests | -0.023 |
Token | upon |
Token ID | 2402 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
pleased | +0.213 |
glad | +0.205 |
ok | +0.199 |
sorry | +0.198 |
okay | +0.196 |
Neg logit contributions | |
board | -0.165 |
x | -0.162 |
els | -0.161 |
oline | -0.161 |
boxes | -0.161 |
Token | a |
Token ID | 257 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
glad | +0.026 |
pleased | +0.025 |
impressed | +0.023 |
sorry | +0.022 |
okay | +0.022 |
Neg logit contributions | |
bons | -0.024 |
growth | -0.022 |
talk | -0.021 |
bars | -0.021 |
ests | -0.019 |
Token | time |
Token ID | 640 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
glad | +0.078 |
pleased | +0.076 |
impressed | +0.071 |
sorry | +0.069 |
lucky | +0.065 |
Neg logit contributions | |
bons | -0.076 |
growth | -0.068 |
bars | -0.066 |
talk | -0.066 |
shirts | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
pleased | +0.247 |
glad | +0.229 |
impressed | +0.227 |
blessed | +0.212 |
thankful | +0.211 |
Neg logit contributions | |
room | -0.090 |
bons | -0.084 |
bars | -0.081 |
ests | -0.071 |
growth | -0.071 |
Token | there |
Token ID | 612 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
pleased | +0.168 |
glad | +0.153 |
impressed | +0.146 |
sorry | +0.136 |
blessed | +0.134 |
Neg logit contributions | |
bons | -0.133 |
bars | -0.133 |
room | -0.130 |
x | -0.130 |
al | -0.128 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
bons | +0.002 |
talk | +0.001 |
growth | +0.001 |
bars | +0.001 |
ests | +0.001 |
Neg logit contributions | |
glad | -0.002 |
pleased | -0.002 |
impressed | -0.001 |
okay | -0.001 |
sorry | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
bons | +0.070 |
talk | +0.065 |
growth | +0.064 |
bars | +0.062 |
ests | +0.060 |
Neg logit contributions | |
glad | -0.071 |
pleased | -0.065 |
lucky | -0.061 |
okay | -0.060 |
sorry | -0.058 |
Token | tw |
Token ID | 665 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
glad | +0.083 |
pleased | +0.075 |
okay | +0.074 |
impressed | +0.073 |
sorry | +0.072 |
Neg logit contributions | |
bons | -0.096 |
bars | -0.085 |
talk | -0.084 |
growth | -0.084 |
fabric | -0.082 |
Token | inkle |
Token ID | 19894 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
glad | +0.084 |
okay | +0.075 |
pleased | +0.075 |
sorry | +0.075 |
impressed | +0.072 |
Neg logit contributions | |
bons | -0.067 |
talk | -0.054 |
growth | -0.054 |
ests | -0.050 |
bars | -0.050 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
glad | +0.033 |
pleased | +0.030 |
impressed | +0.028 |
okay | +0.028 |
sorry | +0.028 |
Neg logit contributions | |
bons | -0.033 |
growth | -0.029 |
talk | -0.028 |
bars | -0.027 |
als | -0.027 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
bons | +0.003 |
talk | +0.003 |
growth | +0.003 |
bars | +0.003 |
ests | +0.003 |
Neg logit contributions | |
glad | -0.003 |
pleased | -0.003 |
impressed | -0.003 |
okay | -0.003 |
sorry | -0.003 |
Token | upon |
Token ID | 2402 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
pleased | +0.210 |
glad | +0.202 |
sorry | +0.197 |
okay | +0.193 |
impressed | +0.191 |
Neg logit contributions | |
boxes | -0.180 |
board | -0.178 |
talk | -0.177 |
ests | -0.174 |
oline | -0.170 |
Token | a |
Token ID | 257 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
glad | +0.051 |
pleased | +0.049 |
impressed | +0.045 |
sorry | +0.044 |
lucky | +0.042 |
Neg logit contributions | |
bons | -0.042 |
growth | -0.038 |
bars | -0.036 |
talk | -0.036 |
ests | -0.033 |
Token | time |
Token ID | 640 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
glad | +0.087 |
pleased | +0.084 |
impressed | +0.079 |
sorry | +0.077 |
lucky | +0.072 |
Neg logit contributions | |
bons | -0.083 |
growth | -0.075 |
talk | -0.073 |
bars | -0.073 |
shirts | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
pleased | +0.267 |
glad | +0.247 |
impressed | +0.246 |
blessed | +0.233 |
thankful | +0.231 |
Neg logit contributions | |
room | -0.099 |
bars | -0.087 |
growth | -0.079 |
bons | -0.078 |
benches | -0.077 |
Token | there |
Token ID | 612 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
pleased | +0.164 |
impressed | +0.148 |
glad | +0.148 |
blessed | +0.128 |
thankful | +0.127 |
Neg logit contributions | |
bars | -0.126 |
growth | -0.125 |
room | -0.124 |
bons | -0.122 |
al | -0.117 |
Token | was |
Token ID | 373 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
glad | +0.053 |
pleased | +0.050 |
impressed | +0.046 |
okay | +0.046 |
sorry | +0.046 |
Neg logit contributions | |
bons | -0.045 |
talk | -0.040 |
bars | -0.040 |
growth | -0.038 |
als | -0.036 |
Token | left |
Token ID | 1364 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
reetings | +0.123 |
pleased | +0.111 |
glad | +0.110 |
mber | +0.104 |
impressed | +0.103 |
Neg logit contributions | |
bons | -0.160 |
bars | -0.156 |
stick | -0.150 |
talk | -0.143 |
growth | -0.131 |
Token | all |
Token ID | 477 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
glad | +0.044 |
pleased | +0.042 |
impressed | +0.038 |
lucky | +0.038 |
okay | +0.037 |
Neg logit contributions | |
bons | -0.050 |
talk | -0.043 |
bars | -0.043 |
growth | -0.042 |
cans | -0.040 |
Token | alone |
Token ID | 3436 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
glad | +0.071 |
pleased | +0.067 |
okay | +0.065 |
impressed | +0.063 |
sorry | +0.062 |
Neg logit contributions | |
bons | -0.075 |
talk | -0.064 |
bars | -0.063 |
cans | -0.060 |
growth | -0.059 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
glad | +0.116 |
pleased | +0.112 |
impressed | +0.103 |
sorry | +0.102 |
lucky | +0.101 |
Neg logit contributions | |
bons | -0.068 |
bars | -0.059 |
talk | -0.056 |
growth | -0.055 |
cans | -0.053 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
glad | +0.015 |
pleased | +0.014 |
impressed | +0.013 |
sorry | +0.013 |
okay | +0.013 |
Neg logit contributions | |
bons | -0.015 |
growth | -0.014 |
talk | -0.014 |
bars | -0.013 |
ests | -0.013 |
Token | there |
Token ID | 612 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
pleased | +0.208 |
glad | +0.192 |
shadow | +0.191 |
impressed | +0.189 |
willing | +0.183 |
Neg logit contributions | |
boxes | -0.177 |
els | -0.175 |
board | -0.174 |
talk | -0.170 |
ens | -0.169 |
Token | was |
Token ID | 373 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
glad | +0.089 |
pleased | +0.086 |
sorry | +0.081 |
impressed | +0.081 |
okay | +0.078 |
Neg logit contributions | |
bons | -0.050 |
growth | -0.043 |
bars | -0.043 |
talk | -0.042 |
ests | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
bons | +0.006 |
growth | +0.005 |
talk | +0.005 |
bars | +0.005 |
shirts | +0.005 |
Neg logit contributions | |
glad | -0.005 |
pleased | -0.005 |
impressed | -0.005 |
okay | -0.004 |
sorry | -0.004 |
Token | little |
Token ID | 1310 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
glad | +0.117 |
pleased | +0.117 |
impressed | +0.105 |
okay | +0.105 |
shadow | +0.103 |
Neg logit contributions | |
bars | -0.117 |
bons | -0.116 |
cash | -0.110 |
shirts | -0.110 |
growth | -0.109 |
Token | girl |
Token ID | 2576 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
glad | +0.026 |
pleased | +0.025 |
impressed | +0.023 |
okay | +0.023 |
lucky | +0.022 |
Neg logit contributions | |
bons | -0.020 |
talk | -0.017 |
bars | -0.017 |
growth | -0.016 |
shirts | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
glad | +0.060 |
pleased | +0.056 |
sorry | +0.053 |
okay | +0.052 |
impressed | +0.052 |
Neg logit contributions | |
bons | -0.046 |
talk | -0.039 |
growth | -0.038 |
shirts | -0.038 |
bars | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
glad | +0.021 |
pleased | +0.020 |
impressed | +0.018 |
okay | +0.018 |
lucky | +0.018 |
Neg logit contributions | |
bons | -0.016 |
growth | -0.014 |
talk | -0.013 |
bars | -0.013 |
shirts | -0.013 |
Token | discovery |
Token ID | 9412 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
glad | +0.058 |
pleased | +0.054 |
impressed | +0.050 |
okay | +0.048 |
sorry | +0.046 |
Neg logit contributions | |
bons | -0.071 |
growth | -0.062 |
shirts | -0.062 |
bars | -0.062 |
talk | -0.060 |
Token | ! |
Token ID | 0 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
bons | +0.055 |
talk | +0.050 |
growth | +0.049 |
bars | +0.046 |
ales | +0.045 |
Neg logit contributions | |
glad | -0.051 |
pleased | -0.048 |
lucky | -0.044 |
impressed | -0.044 |
okay | -0.044 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
glad | +0.021 |
pleased | +0.020 |
impressed | +0.018 |
okay | +0.018 |
lucky | +0.018 |
Neg logit contributions | |
bons | -0.022 |
growth | -0.019 |
talk | -0.019 |
bars | -0.018 |
ests | -0.018 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
glad | +0.005 |
pleased | +0.005 |
impressed | +0.005 |
okay | +0.005 |
sorry | +0.005 |
Neg logit contributions | |
bons | -0.006 |
talk | -0.005 |
growth | -0.005 |
bars | -0.005 |
ests | -0.005 |
Token | upon |
Token ID | 2402 |
Feature activation | +8.38e-01 |
Pos logit contributions | |
pleased | +0.206 |
glad | +0.198 |
sorry | +0.195 |
impressed | +0.188 |
okay | +0.188 |
Neg logit contributions | |
x | -0.173 |
els | -0.172 |
talk | -0.172 |
ests | -0.171 |
board | -0.171 |
Token | a |
Token ID | 257 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
glad | +0.050 |
pleased | +0.048 |
impressed | +0.044 |
sorry | +0.044 |
okay | +0.042 |
Neg logit contributions | |
bons | -0.042 |
growth | -0.038 |
bars | -0.036 |
talk | -0.035 |
shirts | -0.033 |
Token | time |
Token ID | 640 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
glad | +0.104 |
pleased | +0.102 |
impressed | +0.096 |
sorry | +0.094 |
lucky | +0.087 |
Neg logit contributions | |
bons | -0.090 |
growth | -0.080 |
bars | -0.077 |
talk | -0.076 |
shirts | -0.074 |
Token | , |
Token ID | 11 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
pleased | +0.281 |
glad | +0.264 |
impressed | +0.258 |
thankful | +0.247 |
grateful | +0.246 |
Neg logit contributions | |
room | -0.099 |
bars | -0.088 |
bons | -0.082 |
cans | -0.081 |
als | -0.080 |
Token | there |
Token ID | 612 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
pleased | +0.188 |
glad | +0.168 |
impressed | +0.166 |
blessed | +0.152 |
sorry | +0.147 |
Neg logit contributions | |
room | -0.140 |
bars | -0.140 |
x | -0.137 |
growth | -0.137 |
bons | -0.136 |
Token | best |
Token ID | 1266 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
bons | +0.035 |
talk | +0.032 |
growth | +0.032 |
bars | +0.030 |
ales | +0.030 |
Neg logit contributions | |
glad | -0.023 |
pleased | -0.021 |
lucky | -0.019 |
okay | -0.019 |
impressed | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
glad | +0.151 |
pleased | +0.145 |
impressed | +0.137 |
sorry | +0.129 |
okay | +0.125 |
Neg logit contributions | |
bons | -0.152 |
bars | -0.140 |
talk | -0.136 |
shirts | -0.127 |
ests | -0.124 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
glad | +0.081 |
pleased | +0.079 |
impressed | +0.072 |
okay | +0.070 |
lucky | +0.069 |
Neg logit contributions | |
bons | -0.090 |
bars | -0.077 |
talk | -0.074 |
growth | -0.073 |
shirts | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
glad | +0.084 |
pleased | +0.079 |
impressed | +0.073 |
sorry | +0.073 |
lucky | +0.072 |
Neg logit contributions | |
bons | -0.071 |
talk | -0.062 |
bars | -0.060 |
growth | -0.058 |
ales | -0.056 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
bons | +0.021 |
talk | +0.018 |
growth | +0.018 |
bars | +0.017 |
ests | +0.017 |
Neg logit contributions | |
glad | -0.025 |
pleased | -0.023 |
impressed | -0.021 |
okay | -0.021 |
lucky | -0.021 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
glad | +0.013 |
pleased | +0.012 |
impressed | +0.011 |
sorry | +0.011 |
okay | +0.011 |
Neg logit contributions | |
bons | -0.014 |
growth | -0.013 |
talk | -0.013 |
bars | -0.012 |
ests | -0.012 |
Token | upon |
Token ID | 2402 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
pleased | +0.214 |
glad | +0.209 |
sorry | +0.199 |
ok | +0.192 |
impressed | +0.191 |
Neg logit contributions | |
x | -0.172 |
els | -0.172 |
board | -0.169 |
ests | -0.168 |
orns | -0.167 |
Token | a |
Token ID | 257 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
glad | +0.020 |
pleased | +0.019 |
impressed | +0.018 |
sorry | +0.017 |
okay | +0.017 |
Neg logit contributions | |
bons | -0.020 |
growth | -0.018 |
talk | -0.017 |
bars | -0.017 |
ests | -0.016 |
Token | time |
Token ID | 640 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
glad | +0.057 |
pleased | +0.055 |
impressed | +0.051 |
sorry | +0.050 |
lucky | +0.047 |
Neg logit contributions | |
bons | -0.062 |
growth | -0.056 |
talk | -0.055 |
bars | -0.054 |
shirts | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
pleased | +0.248 |
glad | +0.229 |
impressed | +0.226 |
blessed | +0.211 |
thankful | +0.211 |
Neg logit contributions | |
room | -0.096 |
bons | -0.089 |
bars | -0.085 |
cans | -0.077 |
blocks | -0.076 |
Token | there |
Token ID | 612 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
pleased | +0.167 |
glad | +0.155 |
impressed | +0.143 |
sorry | +0.137 |
blessed | +0.132 |
Neg logit contributions | |
bons | -0.141 |
x | -0.138 |
bars | -0.138 |
els | -0.133 |
al | -0.132 |
Token | a |
Token ID | 257 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
bons | +0.020 |
talk | +0.018 |
growth | +0.018 |
bars | +0.017 |
ests | +0.017 |
Neg logit contributions | |
glad | -0.018 |
pleased | -0.017 |
impressed | -0.016 |
okay | -0.016 |
lucky | -0.016 |
Token | special |
Token ID | 2041 |
Feature activation | -4.31e-03 |
Pos logit contributions | |
glad | +0.052 |
pleased | +0.049 |
impressed | +0.046 |
okay | +0.043 |
sorry | +0.042 |
Neg logit contributions | |
bons | -0.081 |
bars | -0.074 |
talk | -0.072 |
shirts | -0.072 |
growth | -0.070 |
Token | moment |
Token ID | 2589 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
bons | +0.000 |
talk | +0.000 |
growth | +0.000 |
bars | +0.000 |
shirts | +0.000 |
Neg logit contributions | |
glad | -0.000 |
pleased | -0.000 |
impressed | -0.000 |
okay | -0.000 |
sorry | -0.000 |
Token | together |
Token ID | 1978 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
glad | +0.033 |
pleased | +0.031 |
impressed | +0.028 |
sorry | +0.028 |
okay | +0.028 |
Neg logit contributions | |
bons | -0.030 |
talk | -0.026 |
growth | -0.025 |
bars | -0.025 |
shirts | -0.024 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
bons | +0.021 |
talk | +0.019 |
growth | +0.019 |
bars | +0.018 |
ests | +0.017 |
Neg logit contributions | |
glad | -0.022 |
pleased | -0.021 |
impressed | -0.019 |
okay | -0.019 |
lucky | -0.019 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
bons | +0.008 |
talk | +0.007 |
growth | +0.007 |
bars | +0.007 |
ests | +0.006 |
Neg logit contributions | |
glad | -0.008 |
pleased | -0.007 |
impressed | -0.006 |
okay | -0.006 |
lucky | -0.006 |
Token | there |
Token ID | 612 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
pleased | +0.209 |
glad | +0.196 |
shadow | +0.193 |
impressed | +0.191 |
sorry | +0.191 |
Neg logit contributions | |
els | -0.170 |
ens | -0.167 |
oline | -0.167 |
boxes | -0.167 |
board | -0.166 |
Token | was |
Token ID | 373 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
glad | +0.090 |
pleased | +0.088 |
sorry | +0.083 |
impressed | +0.083 |
okay | +0.080 |
Neg logit contributions | |
bons | -0.051 |
growth | -0.044 |
bars | -0.043 |
talk | -0.042 |
ests | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
glad | +0.019 |
pleased | +0.019 |
impressed | +0.017 |
sorry | +0.017 |
okay | +0.016 |
Neg logit contributions | |
bons | -0.019 |
growth | -0.017 |
talk | -0.016 |
bars | -0.016 |
shirts | -0.015 |
Token | nos |
Token ID | 43630 |
Feature activation | -9.43e-01 |
Pos logit contributions | |
glad | +0.112 |
pleased | +0.112 |
impressed | +0.101 |
okay | +0.101 |
ok | +0.099 |
Neg logit contributions | |
bons | -0.115 |
bars | -0.112 |
shirts | -0.109 |
cash | -0.107 |
growth | -0.106 |
Token | y |
Token ID | 88 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
bons | +0.054 |
growth | +0.047 |
talk | +0.047 |
bars | +0.045 |
shirts | +0.044 |
Neg logit contributions | |
glad | -0.053 |
pleased | -0.049 |
impressed | -0.045 |
lucky | -0.045 |
okay | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
bons | +0.072 |
talk | +0.066 |
growth | +0.065 |
ales | +0.060 |
ests | +0.060 |
Neg logit contributions | |
glad | -0.062 |
pleased | -0.057 |
okay | -0.054 |
sorry | -0.053 |
impressed | -0.053 |
Token | They |
Token ID | 1119 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
bons | +0.037 |
talk | +0.031 |
growth | +0.030 |
bars | +0.029 |
ales | +0.028 |
Neg logit contributions | |
glad | -0.052 |
pleased | -0.048 |
okay | -0.046 |
sorry | -0.045 |
lucky | -0.045 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
bons | +0.209 |
growth | +0.188 |
ests | +0.182 |
orns | +0.179 |
ales | +0.174 |
Neg logit contributions | |
glad | -0.106 |
pleased | -0.096 |
okay | -0.091 |
happy | -0.090 |
lucky | -0.090 |
Token | around |
Token ID | 1088 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
glad | +0.056 |
pleased | +0.053 |
impressed | +0.048 |
okay | +0.048 |
lucky | +0.046 |
Neg logit contributions | |
bons | -0.056 |
bars | -0.051 |
talk | -0.051 |
shirts | -0.049 |
cans | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
bons | +0.031 |
talk | +0.027 |
growth | +0.027 |
bars | +0.026 |
ests | +0.025 |
Neg logit contributions | |
glad | -0.033 |
pleased | -0.031 |
impressed | -0.028 |
okay | -0.028 |
lucky | -0.028 |
Token | pet |
Token ID | 4273 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
bons | +0.030 |
talk | +0.026 |
growth | +0.026 |
ales | +0.024 |
bars | +0.024 |
Neg logit contributions | |
glad | -0.040 |
pleased | -0.037 |
lucky | -0.035 |
impressed | -0.035 |
okay | -0.034 |
Token | shop |
Token ID | 6128 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
bons | +0.188 |
talk | +0.179 |
growth | +0.165 |
surfaces | +0.148 |
ales | +0.139 |
Neg logit contributions | |
glad | -0.263 |
lucky | -0.240 |
sorry | -0.239 |
happy | -0.236 |
okay | -0.235 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
bons | +0.101 |
growth | +0.092 |
talk | +0.089 |
ests | +0.083 |
ales | +0.081 |
Neg logit contributions | |
glad | -0.079 |
pleased | -0.073 |
lucky | -0.067 |
impressed | -0.067 |
okay | -0.067 |
Token | They |
Token ID | 1119 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
bons | +0.053 |
talk | +0.045 |
growth | +0.041 |
ales | +0.041 |
bars | +0.041 |
Neg logit contributions | |
glad | -0.075 |
pleased | -0.069 |
okay | -0.068 |
sorry | -0.066 |
lucky | -0.066 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
bons | +0.149 |
growth | +0.129 |
ests | +0.126 |
talk | +0.124 |
ales | +0.123 |
Neg logit contributions | |
glad | -0.083 |
pleased | -0.075 |
impressed | -0.070 |
lucky | -0.067 |
okay | -0.067 |
Token | a |
Token ID | 257 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
bons | +0.096 |
talk | +0.088 |
growth | +0.083 |
ales | +0.081 |
ests | +0.079 |
Neg logit contributions | |
glad | -0.072 |
pleased | -0.064 |
sorry | -0.062 |
lucky | -0.061 |
impressed | -0.059 |
Token | call |
Token ID | 869 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
bons | +0.065 |
talk | +0.058 |
growth | +0.056 |
ests | +0.054 |
ales | +0.053 |
Neg logit contributions | |
glad | -0.080 |
pleased | -0.075 |
sorry | -0.071 |
okay | -0.070 |
lucky | -0.070 |
Token | her |
Token ID | 607 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
bons | +0.094 |
growth | +0.088 |
talk | +0.088 |
ests | +0.082 |
bars | +0.080 |
Neg logit contributions | |
glad | -0.082 |
pleased | -0.075 |
okay | -0.074 |
sorry | -0.073 |
lucky | -0.070 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
bons | +0.076 |
talk | +0.072 |
growth | +0.071 |
bars | +0.066 |
ales | +0.066 |
Neg logit contributions | |
glad | -0.045 |
pleased | -0.041 |
lucky | -0.038 |
okay | -0.038 |
impressed | -0.036 |
Token | on |
Token ID | 319 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
bons | +0.085 |
growth | +0.077 |
talk | +0.077 |
ests | +0.070 |
bars | +0.070 |
Neg logit contributions | |
glad | -0.068 |
pleased | -0.062 |
okay | -0.058 |
lucky | -0.057 |
impressed | -0.057 |
Token | her |
Token ID | 607 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
bons | +0.099 |
talk | +0.094 |
ests | +0.086 |
growth | +0.086 |
als | +0.085 |
Neg logit contributions | |
glad | -0.082 |
pleased | -0.074 |
lucky | -0.070 |
impressed | -0.070 |
okay | -0.069 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
bons | +0.078 |
talk | +0.072 |
growth | +0.069 |
ests | +0.065 |
ales | +0.064 |
Neg logit contributions | |
glad | -0.078 |
pleased | -0.071 |
lucky | -0.069 |
okay | -0.067 |
sorry | -0.066 |
Token | phone |
Token ID | 3072 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
bons | +0.265 |
growth | +0.252 |
talk | +0.229 |
aled | +0.221 |
gent | +0.213 |
Neg logit contributions | |
glad | -0.197 |
sorry | -0.185 |
sad | -0.185 |
lucky | -0.180 |
happy | -0.177 |
Token | and |
Token ID | 290 |
Feature activation | -8.90e-01 |
Pos logit contributions | |
bons | +0.198 |
growth | +0.188 |
talk | +0.174 |
ests | +0.166 |
orns | +0.165 |
Neg logit contributions | |
glad | -0.142 |
okay | -0.132 |
pleased | -0.129 |
sorry | -0.123 |
lucky | -0.118 |
Token | help |
Token ID | 1037 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
bons | +0.050 |
talk | +0.044 |
growth | +0.044 |
ests | +0.041 |
bars | +0.041 |
Neg logit contributions | |
glad | -0.051 |
pleased | -0.047 |
impressed | -0.044 |
lucky | -0.044 |
okay | -0.044 |
Token | her |
Token ID | 607 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
bons | +0.130 |
growth | +0.123 |
ests | +0.120 |
talk | +0.118 |
board | +0.112 |
Neg logit contributions | |
glad | -0.097 |
okay | -0.092 |
pleased | -0.089 |
lucky | -0.088 |
happy | -0.087 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
bons | +0.030 |
talk | +0.027 |
growth | +0.026 |
bars | +0.025 |
ests | +0.025 |
Neg logit contributions | |
glad | -0.025 |
pleased | -0.023 |
okay | -0.021 |
lucky | -0.021 |
impressed | -0.021 |
Token | spread |
Token ID | 4104 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
bons | +0.082 |
talk | +0.078 |
ests | +0.071 |
ales | +0.070 |
growth | +0.069 |
Neg logit contributions | |
glad | -0.111 |
pleased | -0.105 |
lucky | -0.100 |
okay | -0.099 |
impressed | -0.098 |
Token | it |
Token ID | 340 |
Feature activation | -8.37e-01 |
Pos logit contributions | |
bons | +0.100 |
talk | +0.093 |
growth | +0.091 |
ests | +0.088 |
ales | +0.086 |
Neg logit contributions | |
glad | -0.070 |
pleased | -0.064 |
okay | -0.060 |
lucky | -0.059 |
sorry | -0.059 |
Token | on |
Token ID | 319 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
bons | +0.054 |
growth | +0.048 |
talk | +0.048 |
ests | +0.045 |
bars | +0.045 |
Neg logit contributions | |
glad | -0.040 |
pleased | -0.037 |
okay | -0.035 |
sorry | -0.034 |
lucky | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
talk | +0.213 |
ests | +0.207 |
bons | +0.206 |
ales | +0.194 |
board | +0.188 |
Neg logit contributions | |
glad | -0.102 |
okay | -0.091 |
sorry | -0.088 |
happy | -0.086 |
lucky | -0.086 |
Token | chew |
Token ID | 34722 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
bons | +0.073 |
talk | +0.068 |
growth | +0.061 |
ests | +0.060 |
ales | +0.060 |
Neg logit contributions | |
glad | -0.122 |
pleased | -0.114 |
lucky | -0.114 |
sorry | -0.112 |
okay | -0.109 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
bons | +0.217 |
talk | +0.209 |
growth | +0.200 |
ales | +0.183 |
stations | +0.181 |
Neg logit contributions | |
okay | -0.117 |
glad | -0.116 |
lucky | -0.112 |
pleased | -0.110 |
sorry | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
bons | +0.312 |
growth | +0.296 |
talk | +0.295 |
aled | +0.283 |
ests | +0.277 |
Neg logit contributions | |
okay | -0.117 |
sorry | -0.110 |
lucky | -0.105 |
happy | -0.100 |
glad | -0.098 |
Token | Then |
Token ID | 3244 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
bons | +0.040 |
talk | +0.036 |
growth | +0.034 |
ales | +0.033 |
bars | +0.032 |
Neg logit contributions | |
glad | -0.049 |
pleased | -0.045 |
okay | -0.043 |
sorry | -0.043 |
lucky | -0.042 |
Token | she |
Token ID | 673 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
glad | +0.019 |
pleased | +0.019 |
impressed | +0.017 |
lucky | +0.016 |
okay | +0.016 |
Neg logit contributions | |
bons | -0.023 |
talk | -0.021 |
bars | -0.021 |
growth | -0.021 |
cans | -0.020 |
Token | went |
Token ID | 1816 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
bons | +0.093 |
talk | +0.080 |
growth | +0.080 |
ales | +0.077 |
ests | +0.075 |
Neg logit contributions | |
glad | -0.083 |
pleased | -0.080 |
impressed | -0.074 |
okay | -0.071 |
sorry | -0.071 |
Token | back |
Token ID | 736 |
Feature activation | -9.07e-01 |
Pos logit contributions | |
bons | +0.029 |
talk | +0.026 |
growth | +0.026 |
bars | +0.025 |
ests | +0.024 |
Neg logit contributions | |
glad | -0.023 |
pleased | -0.021 |
impressed | -0.019 |
lucky | -0.019 |
okay | -0.019 |
Token | They |
Token ID | 2990 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
bons | +0.025 |
talk | +0.022 |
growth | +0.022 |
bars | +0.021 |
ests | +0.021 |
Neg logit contributions | |
glad | -0.019 |
pleased | -0.018 |
impressed | -0.016 |
okay | -0.016 |
lucky | -0.016 |
Token | picked |
Token ID | 6497 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
bons | +0.177 |
orns | +0.151 |
growth | +0.150 |
ests | +0.146 |
ales | +0.146 |
Neg logit contributions | |
glad | -0.122 |
pleased | -0.115 |
impressed | -0.108 |
okay | -0.104 |
sorry | -0.102 |
Token | up |
Token ID | 510 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
bons | +0.161 |
talk | +0.154 |
growth | +0.151 |
ests | +0.148 |
ales | +0.143 |
Neg logit contributions | |
glad | -0.072 |
lucky | -0.060 |
pleased | -0.059 |
sorry | -0.058 |
okay | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
bons | +0.216 |
talk | +0.203 |
growth | +0.196 |
ests | +0.191 |
ales | +0.191 |
Neg logit contributions | |
glad | -0.102 |
lucky | -0.085 |
happy | -0.083 |
sorry | -0.082 |
okay | -0.081 |
Token | fallen |
Token ID | 9292 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
bons | +0.045 |
talk | +0.041 |
growth | +0.040 |
ests | +0.037 |
ales | +0.036 |
Neg logit contributions | |
glad | -0.060 |
pleased | -0.055 |
lucky | -0.053 |
okay | -0.052 |
sorry | -0.051 |
Token | pet |
Token ID | 4273 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
bons | +0.116 |
talk | +0.109 |
growth | +0.102 |
ests | +0.101 |
ales | +0.096 |
Neg logit contributions | |
glad | -0.141 |
pleased | -0.133 |
sorry | -0.133 |
okay | -0.129 |
lucky | -0.127 |
Token | als |
Token ID | 874 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
bons | +0.197 |
talk | +0.178 |
growth | +0.169 |
ests | +0.160 |
ales | +0.154 |
Neg logit contributions | |
glad | -0.236 |
sorry | -0.230 |
okay | -0.224 |
lucky | -0.217 |
happy | -0.213 |
Token | and |
Token ID | 290 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
bons | +0.150 |
talk | +0.134 |
growth | +0.132 |
ests | +0.128 |
ales | +0.121 |
Neg logit contributions | |
glad | -0.124 |
pleased | -0.118 |
okay | -0.111 |
sorry | -0.107 |
impressed | -0.106 |
Token | tried |
Token ID | 3088 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
bons | +0.149 |
ests | +0.138 |
ales | +0.130 |
talk | +0.128 |
growth | +0.125 |
Neg logit contributions | |
glad | -0.120 |
pleased | -0.111 |
sorry | -0.105 |
okay | -0.103 |
happy | -0.103 |
Token | to |
Token ID | 284 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
bons | +0.003 |
talk | +0.003 |
growth | +0.002 |
bars | +0.002 |
shirts | +0.002 |
Neg logit contributions | |
glad | -0.002 |
pleased | -0.002 |
impressed | -0.002 |
lucky | -0.002 |
okay | -0.002 |
Token | put |
Token ID | 1234 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
bons | +0.118 |
ests | +0.114 |
growth | +0.105 |
als | +0.104 |
orns | +0.104 |
Neg logit contributions | |
glad | -0.110 |
okay | -0.107 |
sorry | -0.102 |
pleased | -0.100 |
happy | -0.099 |
Token | back |
Token ID | 736 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
bons | +0.038 |
growth | +0.034 |
talk | +0.034 |
bars | +0.032 |
ests | +0.032 |
Neg logit contributions | |
glad | -0.028 |
pleased | -0.026 |
impressed | -0.023 |
lucky | -0.023 |
sorry | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
bons | +0.111 |
growth | +0.101 |
talk | +0.099 |
ests | +0.095 |
ales | +0.094 |
Neg logit contributions | |
glad | -0.057 |
pleased | -0.049 |
lucky | -0.045 |
impressed | -0.043 |
happy | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
bons | +0.041 |
talk | +0.038 |
growth | +0.036 |
bars | +0.035 |
ests | +0.035 |
Neg logit contributions | |
glad | -0.037 |
pleased | -0.034 |
impressed | -0.031 |
sorry | -0.031 |
lucky | -0.031 |
Token | pile |
Token ID | 14540 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
bons | +0.033 |
talk | +0.029 |
growth | +0.029 |
ales | +0.026 |
bars | +0.026 |
Neg logit contributions | |
glad | -0.050 |
pleased | -0.047 |
lucky | -0.044 |
impressed | -0.044 |
sorry | -0.044 |
Token | of |
Token ID | 286 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
bons | +0.228 |
growth | +0.206 |
talk | +0.197 |
ales | +0.186 |
ests | +0.181 |
Neg logit contributions | |
glad | -0.129 |
pleased | -0.122 |
impressed | -0.108 |
happy | -0.106 |
sorry | -0.104 |
Token | ash |
Token ID | 12530 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
talk | +0.077 |
bons | +0.074 |
growth | +0.069 |
ests | +0.061 |
als | +0.060 |
Neg logit contributions | |
glad | -0.133 |
pleased | -0.125 |
impressed | -0.122 |
lucky | -0.121 |
happy | -0.120 |
Token | and |
Token ID | 290 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
bons | +0.259 |
growth | +0.251 |
Mine | +0.240 |
talk | +0.226 |
Warning | +0.223 |
Neg logit contributions | |
glad | -0.144 |
happy | -0.135 |
sorry | -0.130 |
okay | -0.129 |
sad | -0.129 |
Token | stirred |
Token ID | 33091 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
bons | +0.120 |
ests | +0.107 |
talk | +0.105 |
growth | +0.104 |
ales | +0.100 |
Neg logit contributions | |
glad | -0.106 |
pleased | -0.099 |
impressed | -0.093 |
sorry | -0.090 |
happy | -0.088 |
Token | it |
Token ID | 340 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
bons | +0.109 |
talk | +0.106 |
growth | +0.101 |
ests | +0.097 |
ales | +0.095 |
Neg logit contributions | |
glad | -0.069 |
pleased | -0.064 |
okay | -0.058 |
lucky | -0.058 |
impressed | -0.058 |
Token | with |
Token ID | 351 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
bons | +0.095 |
growth | +0.088 |
talk | +0.088 |
ests | +0.083 |
ales | +0.081 |
Neg logit contributions | |
glad | -0.067 |
pleased | -0.064 |
okay | -0.059 |
sorry | -0.057 |
impressed | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
talk | +0.123 |
bons | +0.123 |
ests | +0.112 |
ales | +0.110 |
growth | +0.109 |
Neg logit contributions | |
glad | -0.095 |
pleased | -0.086 |
sorry | -0.081 |
lucky | -0.080 |
okay | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
bons | +0.095 |
talk | +0.089 |
growth | +0.088 |
ales | +0.086 |
ests | +0.086 |
Neg logit contributions | |
glad | -0.062 |
pleased | -0.056 |
lucky | -0.055 |
sorry | -0.053 |
okay | -0.053 |
Token | ball |
Token ID | 2613 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
bons | +0.041 |
talk | +0.037 |
growth | +0.036 |
ales | +0.033 |
ests | +0.031 |
Neg logit contributions | |
glad | -0.060 |
pleased | -0.057 |
lucky | -0.054 |
okay | -0.053 |
sorry | -0.053 |
Token | out |
Token ID | 503 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
bons | +0.203 |
growth | +0.188 |
ests | +0.176 |
talk | +0.171 |
occasions | +0.166 |
Neg logit contributions | |
glad | -0.124 |
happy | -0.111 |
pleased | -0.109 |
okay | -0.107 |
sad | -0.107 |
Token | of |
Token ID | 286 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
bons | +0.076 |
talk | +0.068 |
growth | +0.067 |
ests | +0.064 |
ales | +0.064 |
Neg logit contributions | |
glad | -0.054 |
pleased | -0.051 |
impressed | -0.046 |
sorry | -0.045 |
okay | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
talk | +0.153 |
bons | +0.148 |
als | +0.139 |
ests | +0.139 |
ales | +0.137 |
Neg logit contributions | |
glad | -0.083 |
pleased | -0.070 |
impressed | -0.066 |
happy | -0.064 |
lucky | -0.063 |
Token | cup |
Token ID | 6508 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
bons | +0.091 |
growth | +0.090 |
talk | +0.088 |
ales | +0.088 |
ests | +0.076 |
Neg logit contributions | |
glad | -0.103 |
lucky | -0.096 |
sorry | -0.093 |
pleased | -0.090 |
sad | -0.087 |
Token | board |
Token ID | 3526 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
bons | +0.262 |
growth | +0.249 |
ales | +0.223 |
talk | +0.221 |
Mine | +0.217 |
Neg logit contributions | |
glad | -0.171 |
pleased | -0.157 |
sorry | -0.156 |
happy | -0.155 |
sad | -0.145 |
Token | and |
Token ID | 290 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
bons | +0.169 |
growth | +0.158 |
talk | +0.148 |
ales | +0.145 |
ests | +0.143 |
Neg logit contributions | |
glad | -0.133 |
pleased | -0.125 |
impressed | -0.115 |
sorry | -0.114 |
happy | -0.113 |
Token | closed |
Token ID | 4838 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
bons | +0.109 |
ests | +0.097 |
growth | +0.097 |
talk | +0.096 |
ales | +0.093 |
Neg logit contributions | |
glad | -0.092 |
pleased | -0.087 |
impressed | -0.082 |
sorry | -0.079 |
lucky | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
bons | +0.046 |
talk | +0.042 |
growth | +0.041 |
bars | +0.040 |
ests | +0.039 |
Neg logit contributions | |
glad | -0.035 |
pleased | -0.032 |
impressed | -0.029 |
okay | -0.029 |
sorry | -0.029 |
Token | door |
Token ID | 3420 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
bons | +0.113 |
growth | +0.106 |
talk | +0.105 |
ales | +0.104 |
ests | +0.099 |
Neg logit contributions | |
glad | -0.105 |
sorry | -0.098 |
lucky | -0.096 |
pleased | -0.094 |
sad | -0.087 |
Token |
|
Token ID | 198 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
bons | +0.009 |
talk | +0.009 |
growth | +0.008 |
bars | +0.008 |
ests | +0.008 |
Neg logit contributions | |
glad | -0.007 |
pleased | -0.007 |
impressed | -0.006 |
okay | -0.006 |
lucky | -0.006 |
Token | They |
Token ID | 2990 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
bons | +0.076 |
growth | +0.070 |
talk | +0.066 |
als | +0.065 |
ales | +0.065 |
Neg logit contributions | |
glad | -0.076 |
pleased | -0.071 |
impressed | -0.067 |
okay | -0.065 |
lucky | -0.063 |
Token | went |
Token ID | 1816 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
bons | +0.193 |
orns | +0.166 |
growth | +0.164 |
ests | +0.164 |
ales | +0.156 |
Neg logit contributions | |
glad | -0.134 |
pleased | -0.125 |
impressed | -0.117 |
happy | -0.114 |
okay | -0.112 |
Token | to |
Token ID | 284 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
bons | +0.023 |
talk | +0.021 |
growth | +0.021 |
bars | +0.020 |
ests | +0.019 |
Neg logit contributions | |
glad | -0.018 |
pleased | -0.017 |
impressed | -0.015 |
okay | -0.015 |
lucky | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
bons | +0.109 |
talk | +0.099 |
ests | +0.098 |
growth | +0.093 |
als | +0.093 |
Neg logit contributions | |
glad | -0.074 |
pleased | -0.063 |
lucky | -0.060 |
sorry | -0.060 |
okay | -0.058 |
Token | pet |
Token ID | 4273 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
bons | +0.059 |
talk | +0.052 |
growth | +0.051 |
ales | +0.048 |
ests | +0.046 |
Neg logit contributions | |
glad | -0.093 |
pleased | -0.086 |
lucky | -0.085 |
sorry | -0.082 |
okay | -0.079 |
Token | shop |
Token ID | 6128 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
bons | +0.232 |
growth | +0.197 |
talk | +0.194 |
surfaces | +0.179 |
ales | +0.173 |
Neg logit contributions | |
glad | -0.223 |
sorry | -0.201 |
pleased | -0.195 |
lucky | -0.193 |
happy | -0.187 |
Token | and |
Token ID | 290 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
bons | +0.115 |
growth | +0.105 |
talk | +0.096 |
ests | +0.094 |
orns | +0.091 |
Neg logit contributions | |
glad | -0.094 |
pleased | -0.087 |
impressed | -0.080 |
lucky | -0.079 |
okay | -0.077 |
Token | bought |
Token ID | 5839 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
bons | +0.113 |
ests | +0.098 |
growth | +0.095 |
talk | +0.094 |
orns | +0.092 |
Neg logit contributions | |
glad | -0.091 |
pleased | -0.083 |
lucky | -0.076 |
impressed | -0.076 |
sorry | -0.075 |
Token | a |
Token ID | 257 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
bons | +0.110 |
talk | +0.104 |
growth | +0.102 |
ests | +0.096 |
ales | +0.091 |
Neg logit contributions | |
glad | -0.101 |
pleased | -0.091 |
lucky | -0.090 |
sorry | -0.086 |
okay | -0.086 |
Token | box |
Token ID | 3091 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
glad | +0.050 |
pleased | +0.046 |
impressed | +0.044 |
okay | +0.043 |
sorry | +0.043 |
Neg logit contributions | |
bons | -0.032 |
bars | -0.029 |
talk | -0.028 |
shirts | -0.026 |
growth | -0.026 |
Token | They |
Token ID | 1119 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
bons | +0.081 |
growth | +0.075 |
talk | +0.074 |
ales | +0.071 |
ests | +0.070 |
Neg logit contributions | |
glad | -0.057 |
pleased | -0.051 |
okay | -0.047 |
sorry | -0.047 |
lucky | -0.045 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
bons | +0.165 |
ests | +0.144 |
growth | +0.143 |
orns | +0.142 |
talk | +0.134 |
Neg logit contributions | |
glad | -0.134 |
pleased | -0.126 |
impressed | -0.119 |
happy | -0.118 |
okay | -0.117 |
Token | through |
Token ID | 832 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
glad | +0.016 |
pleased | +0.015 |
okay | +0.014 |
impressed | +0.014 |
sorry | +0.013 |
Neg logit contributions | |
bons | -0.016 |
talk | -0.014 |
growth | -0.013 |
bars | -0.013 |
shirts | -0.013 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
bons | +0.188 |
talk | +0.182 |
ests | +0.179 |
ales | +0.172 |
board | +0.171 |
Neg logit contributions | |
glad | -0.105 |
lucky | -0.093 |
pleased | -0.091 |
happy | -0.091 |
impressed | -0.086 |
Token | 's |
Token ID | 338 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
bons | +0.098 |
talk | +0.091 |
growth | +0.090 |
ests | +0.087 |
ales | +0.081 |
Neg logit contributions | |
glad | -0.085 |
pleased | -0.078 |
lucky | -0.074 |
okay | -0.073 |
impressed | -0.072 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
bons | +0.093 |
talk | +0.091 |
growth | +0.083 |
ests | +0.080 |
ales | +0.079 |
Neg logit contributions | |
glad | -0.122 |
lucky | -0.109 |
pleased | -0.108 |
sorry | -0.104 |
happy | -0.101 |
Token | box |
Token ID | 3091 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
bons | +0.231 |
growth | +0.219 |
talk | +0.210 |
aled | +0.196 |
oline | +0.193 |
Neg logit contributions | |
glad | -0.188 |
lucky | -0.179 |
sad | -0.175 |
happy | -0.174 |
okay | -0.167 |
Token | and |
Token ID | 290 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
bons | +0.106 |
growth | +0.097 |
talk | +0.096 |
ests | +0.089 |
bars | +0.086 |
Neg logit contributions | |
glad | -0.098 |
pleased | -0.091 |
okay | -0.086 |
impressed | -0.084 |
lucky | -0.083 |
Token | found |
Token ID | 1043 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
bons | +0.039 |
talk | +0.035 |
growth | +0.035 |
bars | +0.033 |
ests | +0.033 |
Neg logit contributions | |
glad | -0.033 |
pleased | -0.031 |
impressed | -0.028 |
okay | -0.028 |
lucky | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
bons | +0.075 |
talk | +0.069 |
growth | +0.066 |
ests | +0.064 |
shirts | +0.063 |
Neg logit contributions | |
glad | -0.061 |
pleased | -0.054 |
lucky | -0.052 |
sorry | -0.051 |
okay | -0.050 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
glad | +0.104 |
pleased | +0.097 |
impressed | +0.089 |
okay | +0.088 |
sorry | +0.087 |
Neg logit contributions | |
bons | -0.082 |
bars | -0.076 |
cans | -0.069 |
shirts | -0.069 |
boxes | -0.068 |
Token | ." |
Token ID | 526 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
glad | +0.002 |
pleased | +0.002 |
impressed | +0.002 |
lucky | +0.002 |
sorry | +0.002 |
Neg logit contributions | |
bons | -0.002 |
talk | -0.002 |
growth | -0.002 |
bars | -0.002 |
shirts | -0.002 |
Token | They |
Token ID | 1119 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
bons | +0.098 |
growth | +0.085 |
ales | +0.080 |
talk | +0.080 |
ests | +0.076 |
Neg logit contributions | |
glad | -0.112 |
pleased | -0.097 |
sorry | -0.094 |
lucky | -0.091 |
okay | -0.090 |
Token | take |
Token ID | 1011 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
bons | +0.145 |
growth | +0.133 |
ests | +0.132 |
orns | +0.126 |
talk | +0.125 |
Neg logit contributions | |
glad | -0.134 |
pleased | -0.121 |
sorry | -0.117 |
happy | -0.115 |
impressed | -0.114 |
Token | out |
Token ID | 503 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
bons | +0.147 |
growth | +0.134 |
ests | +0.134 |
ales | +0.132 |
talk | +0.131 |
Neg logit contributions | |
glad | -0.121 |
lucky | -0.112 |
sorry | -0.109 |
pleased | -0.105 |
happy | -0.104 |
Token | their |
Token ID | 511 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
bons | +0.155 |
talk | +0.141 |
growth | +0.136 |
ales | +0.135 |
ests | +0.135 |
Neg logit contributions | |
glad | -0.090 |
pleased | -0.078 |
happy | -0.074 |
lucky | -0.073 |
impressed | -0.071 |
Token | file |
Token ID | 2393 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
bons | +0.079 |
talk | +0.077 |
growth | +0.072 |
ests | +0.064 |
ales | +0.061 |
Neg logit contributions | |
glad | -0.161 |
lucky | -0.151 |
pleased | -0.149 |
sorry | -0.144 |
impressed | -0.141 |
Token | and |
Token ID | 290 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
bons | +0.290 |
growth | +0.290 |
ales | +0.259 |
talk | +0.252 |
Mine | +0.252 |
Neg logit contributions | |
glad | -0.150 |
happy | -0.139 |
pleased | -0.135 |
okay | -0.134 |
sad | -0.129 |
Token | their |
Token ID | 511 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
bons | +0.151 |
ests | +0.148 |
talk | +0.135 |
growth | +0.134 |
orns | +0.132 |
Neg logit contributions | |
glad | -0.156 |
sorry | -0.147 |
pleased | -0.141 |
okay | -0.136 |
happy | -0.136 |
Token | cr |
Token ID | 1067 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
bons | +0.109 |
talk | +0.104 |
growth | +0.101 |
ales | +0.092 |
ests | +0.089 |
Neg logit contributions | |
glad | -0.152 |
sorry | -0.138 |
pleased | -0.137 |
lucky | -0.137 |
impressed | -0.129 |
Token | ay |
Token ID | 323 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
growth | +0.150 |
bons | +0.147 |
talk | +0.143 |
mats | +0.142 |
bars | +0.141 |
Neg logit contributions | |
glad | -0.146 |
pleased | -0.137 |
okay | -0.131 |
impressed | -0.122 |
sorry | -0.121 |
Token | ons |
Token ID | 684 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
growth | +0.099 |
bons | +0.097 |
talk | +0.092 |
bars | +0.084 |
stations | +0.082 |
Neg logit contributions | |
glad | -0.140 |
pleased | -0.129 |
lucky | -0.121 |
okay | -0.119 |
sorry | -0.119 |
Token | idea |
Token ID | 2126 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
bons | +0.074 |
talk | +0.069 |
growth | +0.068 |
ests | +0.064 |
als | +0.064 |
Neg logit contributions | |
glad | -0.056 |
pleased | -0.052 |
lucky | -0.048 |
sorry | -0.047 |
okay | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
bons | +0.036 |
talk | +0.032 |
growth | +0.031 |
bars | +0.030 |
ests | +0.029 |
Neg logit contributions | |
glad | -0.035 |
pleased | -0.033 |
okay | -0.030 |
impressed | -0.030 |
lucky | -0.030 |
Token | She |
Token ID | 1375 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
bons | +0.021 |
talk | +0.018 |
growth | +0.018 |
bars | +0.017 |
ales | +0.016 |
Neg logit contributions | |
glad | -0.028 |
pleased | -0.027 |
okay | -0.025 |
impressed | -0.025 |
lucky | -0.025 |
Token | took |
Token ID | 1718 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
bons | +0.034 |
talk | +0.029 |
growth | +0.029 |
bars | +0.028 |
ests | +0.027 |
Neg logit contributions | |
glad | -0.035 |
pleased | -0.033 |
impressed | -0.030 |
okay | -0.030 |
lucky | -0.030 |
Token | her |
Token ID | 607 |
Feature activation | -9.59e-01 |
Pos logit contributions | |
bons | +0.156 |
ests | +0.149 |
talk | +0.149 |
ales | +0.146 |
growth | +0.143 |
Neg logit contributions | |
glad | -0.084 |
okay | -0.075 |
sorry | -0.075 |
lucky | -0.075 |
pleased | -0.072 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
bons | +0.039 |
talk | +0.034 |
growth | +0.033 |
ales | +0.030 |
ests | +0.030 |
Neg logit contributions | |
glad | -0.070 |
pleased | -0.066 |
lucky | -0.063 |
okay | -0.062 |
sorry | -0.062 |
Token | soldiers |
Token ID | 5795 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
bons | +0.264 |
growth | +0.251 |
talk | +0.229 |
ales | +0.222 |
stations | +0.217 |
Neg logit contributions | |
glad | -0.149 |
sorry | -0.144 |
lucky | -0.143 |
sad | -0.138 |
happy | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
bons | +0.085 |
growth | +0.075 |
talk | +0.074 |
ests | +0.071 |
bars | +0.069 |
Neg logit contributions | |
glad | -0.084 |
pleased | -0.078 |
okay | -0.072 |
impressed | -0.072 |
lucky | -0.072 |
Token | made |
Token ID | 925 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
bons | +0.141 |
growth | +0.126 |
talk | +0.125 |
ests | +0.125 |
ales | +0.118 |
Neg logit contributions | |
glad | -0.125 |
pleased | -0.119 |
impressed | -0.114 |
lucky | -0.112 |
sorry | -0.111 |
Token | them |
Token ID | 606 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
bons | +0.034 |
talk | +0.031 |
growth | +0.030 |
bars | +0.029 |
ests | +0.028 |
Neg logit contributions | |
glad | -0.038 |
pleased | -0.036 |
impressed | -0.033 |
lucky | -0.033 |
okay | -0.033 |
Token | march |
Token ID | 9960 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
bons | +0.036 |
talk | +0.032 |
growth | +0.032 |
bars | +0.029 |
ests | +0.029 |
Neg logit contributions | |
glad | -0.041 |
pleased | -0.038 |
impressed | -0.035 |
okay | -0.035 |
lucky | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
bons | +0.072 |
talk | +0.067 |
growth | +0.064 |
ales | +0.062 |
shirts | +0.062 |
Neg logit contributions | |
glad | -0.050 |
pleased | -0.043 |
lucky | -0.042 |
okay | -0.041 |
sorry | -0.041 |
Token | big |
Token ID | 1263 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
glad | +0.101 |
pleased | +0.095 |
impressed | +0.087 |
okay | +0.081 |
sorry | +0.080 |
Neg logit contributions | |
bons | -0.096 |
bars | -0.091 |
paper | -0.088 |
tape | -0.087 |
cans | -0.087 |
Token | ball |
Token ID | 2613 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
bons | +0.018 |
talk | +0.016 |
growth | +0.016 |
bars | +0.015 |
ests | +0.014 |
Neg logit contributions | |
glad | -0.024 |
pleased | -0.022 |
impressed | -0.021 |
okay | -0.021 |
lucky | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
bons | +0.109 |
talk | +0.100 |
growth | +0.099 |
ests | +0.093 |
ales | +0.092 |
Neg logit contributions | |
glad | -0.067 |
pleased | -0.058 |
sorry | -0.056 |
lucky | -0.054 |
okay | -0.053 |
Token | brilliant |
Token ID | 10457 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
talk | +0.071 |
bons | +0.062 |
growth | +0.059 |
ales | +0.051 |
ests | +0.048 |
Neg logit contributions | |
glad | -0.154 |
pleased | -0.141 |
sorry | -0.141 |
lucky | -0.140 |
okay | -0.139 |
Token | red |
Token ID | 2266 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
talk | +0.109 |
growth | +0.100 |
bons | +0.096 |
ales | +0.090 |
ests | +0.084 |
Neg logit contributions | |
glad | -0.130 |
pleased | -0.125 |
lucky | -0.125 |
sorry | -0.119 |
sad | -0.119 |
Token | yarn |
Token ID | 21181 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
talk | +0.228 |
bons | +0.211 |
growth | +0.206 |
ales | +0.194 |
x | +0.179 |
Neg logit contributions | |
okay | -0.168 |
sad | -0.164 |
sorry | -0.157 |
glad | -0.149 |
there | -0.148 |
Token | in |
Token ID | 287 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
bons | +0.139 |
talk | +0.131 |
growth | +0.127 |
ales | +0.119 |
ests | +0.117 |
Neg logit contributions | |
glad | -0.088 |
sorry | -0.077 |
pleased | -0.076 |
lucky | -0.075 |
okay | -0.075 |
Token | a |
Token ID | 257 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
talk | +0.188 |
bons | +0.185 |
als | +0.178 |
ests | +0.177 |
board | +0.176 |
Neg logit contributions | |
glad | -0.085 |
lucky | -0.073 |
happy | -0.070 |
pleased | -0.064 |
impressed | -0.060 |
Token | box |
Token ID | 3091 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
bons | +0.013 |
talk | +0.012 |
growth | +0.011 |
bars | +0.011 |
shirts | +0.010 |
Neg logit contributions | |
glad | -0.016 |
pleased | -0.015 |
impressed | -0.014 |
okay | -0.014 |
lucky | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
bons | +0.062 |
growth | +0.055 |
talk | +0.055 |
ests | +0.051 |
bars | +0.051 |
Neg logit contributions | |
glad | -0.056 |
pleased | -0.052 |
okay | -0.048 |
impressed | -0.048 |
lucky | -0.047 |
Token | They |
Token ID | 2990 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
bons | +0.056 |
talk | +0.049 |
growth | +0.049 |
bars | +0.047 |
ales | +0.046 |
Neg logit contributions | |
glad | -0.059 |
pleased | -0.056 |
impressed | -0.053 |
lucky | -0.050 |
okay | -0.050 |
Token | went |
Token ID | 1816 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
bons | +0.168 |
growth | +0.143 |
ests | +0.138 |
orns | +0.137 |
talk | +0.136 |
Neg logit contributions | |
glad | -0.125 |
pleased | -0.119 |
impressed | -0.112 |
okay | -0.105 |
happy | -0.105 |
Token | to |
Token ID | 284 |
Feature activation | -9.80e-01 |
Pos logit contributions | |
bons | +0.027 |
growth | +0.024 |
talk | +0.024 |
bars | +0.023 |
ests | +0.022 |
Neg logit contributions | |
glad | -0.021 |
pleased | -0.019 |
impressed | -0.017 |
okay | -0.017 |
lucky | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
bons | +0.062 |
talk | +0.056 |
ests | +0.054 |
growth | +0.053 |
bars | +0.052 |
Neg logit contributions | |
glad | -0.051 |
pleased | -0.046 |
sorry | -0.042 |
okay | -0.042 |
lucky | -0.042 |
Token | nearby |
Token ID | 6716 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
bons | +0.048 |
talk | +0.042 |
growth | +0.041 |
ales | +0.039 |
ests | +0.038 |
Neg logit contributions | |
glad | -0.067 |
pleased | -0.062 |
lucky | -0.059 |
sorry | -0.058 |
impressed | -0.058 |
Token | recycling |
Token ID | 25914 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
bons | +0.167 |
talk | +0.142 |
ests | +0.138 |
growth | +0.137 |
els | +0.135 |
Neg logit contributions | |
glad | -0.180 |
pleased | -0.166 |
happy | -0.160 |
lucky | -0.160 |
impressed | -0.153 |
Token | bin |
Token ID | 9874 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
bons | +0.196 |
talk | +0.191 |
ales | +0.183 |
growth | +0.177 |
ests | +0.171 |
Neg logit contributions | |
sorry | -0.203 |
glad | -0.201 |
sad | -0.197 |
lucky | -0.187 |
happy | -0.186 |
Token | and |
Token ID | 290 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
bons | +0.227 |
growth | +0.215 |
talk | +0.187 |
ests | +0.184 |
Mine | +0.184 |
Neg logit contributions | |
glad | -0.138 |
pleased | -0.121 |
impressed | -0.115 |
happy | -0.112 |
sad | -0.109 |
Token | found |
Token ID | 1043 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
bons | +0.111 |
ests | +0.096 |
growth | +0.095 |
talk | +0.093 |
orns | +0.091 |
Neg logit contributions | |
glad | -0.090 |
pleased | -0.083 |
okay | -0.076 |
lucky | -0.076 |
impressed | -0.076 |
Token | some |
Token ID | 617 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
bons | +0.087 |
talk | +0.080 |
growth | +0.076 |
ests | +0.074 |
ales | +0.072 |
Neg logit contributions | |
glad | -0.069 |
pleased | -0.060 |
sorry | -0.059 |
lucky | -0.059 |
okay | -0.057 |
Token | cardboard |
Token ID | 26819 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
bons | +0.072 |
talk | +0.068 |
growth | +0.066 |
ales | +0.057 |
ests | +0.056 |
Neg logit contributions | |
glad | -0.117 |
pleased | -0.111 |
sorry | -0.106 |
lucky | -0.106 |
okay | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
bons | +0.091 |
growth | +0.080 |
talk | +0.078 |
ales | +0.077 |
bars | +0.074 |
Neg logit contributions | |
glad | -0.066 |
pleased | -0.058 |
sorry | -0.053 |
impressed | -0.052 |
okay | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
bons | +0.002 |
talk | +0.002 |
growth | +0.002 |
bars | +0.001 |
ests | +0.001 |
Neg logit contributions | |
glad | -0.001 |
pleased | -0.001 |
okay | -0.001 |
impressed | -0.001 |
lucky | -0.001 |
Token | They |
Token ID | 2990 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
bons | +0.062 |
growth | +0.057 |
talk | +0.054 |
ales | +0.053 |
als | +0.052 |
Neg logit contributions | |
glad | -0.062 |
pleased | -0.058 |
impressed | -0.055 |
lucky | -0.051 |
okay | -0.051 |
Token | take |
Token ID | 1011 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
bons | +0.145 |
growth | +0.131 |
ests | +0.129 |
orns | +0.126 |
ales | +0.123 |
Neg logit contributions | |
glad | -0.142 |
pleased | -0.131 |
sorry | -0.127 |
happy | -0.123 |
lucky | -0.122 |
Token | the |
Token ID | 262 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
bons | +0.170 |
ests | +0.160 |
ales | +0.156 |
growth | +0.155 |
talk | +0.155 |
Neg logit contributions | |
glad | -0.116 |
lucky | -0.108 |
sorry | -0.106 |
happy | -0.105 |
okay | -0.103 |
Token | chalk |
Token ID | 30860 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
bons | +0.037 |
talk | +0.033 |
growth | +0.033 |
bars | +0.031 |
ests | +0.030 |
Neg logit contributions | |
glad | -0.043 |
pleased | -0.041 |
lucky | -0.038 |
impressed | -0.038 |
okay | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -3.30e+00 |
Pos logit contributions | |
bons | +0.288 |
growth | +0.270 |
ales | +0.255 |
ests | +0.243 |
talk | +0.241 |
Neg logit contributions | |
glad | -0.144 |
happy | -0.140 |
pleased | -0.138 |
okay | -0.133 |
sorry | -0.131 |
Token | draw |
Token ID | 3197 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
bons | +0.180 |
ests | +0.176 |
orns | +0.167 |
ales | +0.159 |
talk | +0.159 |
Neg logit contributions | |
glad | -0.170 |
sorry | -0.161 |
happy | -0.154 |
pleased | -0.154 |
okay | -0.146 |
Token | on |
Token ID | 319 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
bons | +0.106 |
talk | +0.100 |
growth | +0.098 |
ales | +0.095 |
ests | +0.090 |
Neg logit contributions | |
glad | -0.103 |
pleased | -0.099 |
impressed | -0.092 |
sorry | -0.091 |
okay | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
talk | +0.197 |
bons | +0.195 |
ests | +0.188 |
orns | +0.180 |
ales | +0.179 |
Neg logit contributions | |
glad | -0.104 |
happy | -0.092 |
pleased | -0.092 |
impressed | -0.089 |
lucky | -0.087 |
Token | black |
Token ID | 2042 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
bons | +0.077 |
talk | +0.071 |
growth | +0.069 |
ests | +0.065 |
ales | +0.064 |
Neg logit contributions | |
glad | -0.101 |
pleased | -0.094 |
lucky | -0.091 |
sorry | -0.089 |
impressed | -0.087 |
Token | push |
Token ID | 4574 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
bons | +0.014 |
talk | +0.012 |
growth | +0.012 |
bars | +0.012 |
shirts | +0.011 |
Neg logit contributions | |
glad | -0.013 |
pleased | -0.012 |
impressed | -0.011 |
okay | -0.011 |
lucky | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
bons | +0.117 |
growth | +0.111 |
talk | +0.108 |
ests | +0.102 |
ales | +0.102 |
Neg logit contributions | |
glad | -0.048 |
pleased | -0.043 |
sorry | -0.041 |
okay | -0.040 |
lucky | -0.036 |
Token | they |
Token ID | 484 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
bons | +0.012 |
talk | +0.010 |
growth | +0.010 |
bars | +0.010 |
shirts | +0.010 |
Neg logit contributions | |
glad | -0.010 |
pleased | -0.010 |
impressed | -0.009 |
okay | -0.009 |
lucky | -0.009 |
Token | opened |
Token ID | 4721 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
bons | +0.123 |
growth | +0.110 |
ests | +0.101 |
talk | +0.100 |
ales | +0.098 |
Neg logit contributions | |
glad | -0.100 |
pleased | -0.093 |
okay | -0.087 |
impressed | -0.087 |
lucky | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
bons | +0.100 |
talk | +0.095 |
ests | +0.092 |
growth | +0.091 |
ales | +0.088 |
Neg logit contributions | |
glad | -0.066 |
pleased | -0.058 |
lucky | -0.054 |
okay | -0.054 |
sorry | -0.053 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
bons | +0.060 |
talk | +0.053 |
growth | +0.052 |
ales | +0.051 |
ests | +0.047 |
Neg logit contributions | |
glad | -0.087 |
lucky | -0.081 |
sorry | -0.080 |
pleased | -0.080 |
okay | -0.077 |
Token | box |
Token ID | 3091 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
bons | +0.271 |
growth | +0.242 |
talk | +0.238 |
ales | +0.237 |
gent | +0.229 |
Neg logit contributions | |
glad | -0.145 |
lucky | -0.138 |
sorry | -0.137 |
there | -0.134 |
happy | -0.125 |
Token | together |
Token ID | 1978 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
bons | +0.157 |
growth | +0.145 |
talk | +0.137 |
ests | +0.135 |
orns | +0.130 |
Neg logit contributions | |
glad | -0.118 |
pleased | -0.111 |
okay | -0.103 |
impressed | -0.102 |
lucky | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
bons | +0.078 |
talk | +0.071 |
growth | +0.070 |
ests | +0.067 |
ales | +0.064 |
Neg logit contributions | |
glad | -0.064 |
pleased | -0.060 |
okay | -0.055 |
impressed | -0.054 |
lucky | -0.054 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
bons | +0.063 |
talk | +0.057 |
growth | +0.054 |
ales | +0.053 |
bars | +0.051 |
Neg logit contributions | |
glad | -0.065 |
pleased | -0.060 |
sorry | -0.057 |
okay | -0.056 |
impressed | -0.055 |
Token | was |
Token ID | 373 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
bons | +0.071 |
growth | +0.064 |
talk | +0.063 |
ales | +0.061 |
ests | +0.059 |
Neg logit contributions | |
glad | -0.060 |
pleased | -0.057 |
impressed | -0.053 |
lucky | -0.051 |
okay | -0.050 |
Token | sees |
Token ID | 7224 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
bons | +0.050 |
talk | +0.044 |
growth | +0.043 |
ales | +0.041 |
ests | +0.041 |
Neg logit contributions | |
glad | -0.058 |
pleased | -0.054 |
impressed | -0.051 |
lucky | -0.050 |
sorry | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
bons | +0.103 |
talk | +0.097 |
growth | +0.091 |
ales | +0.090 |
ests | +0.088 |
Neg logit contributions | |
glad | -0.070 |
pleased | -0.063 |
lucky | -0.059 |
sorry | -0.059 |
impressed | -0.058 |
Token | lock |
Token ID | 5793 |
Feature activation | -9.15e-01 |
Pos logit contributions | |
glad | +0.079 |
pleased | +0.073 |
okay | +0.068 |
impressed | +0.067 |
sorry | +0.067 |
Neg logit contributions | |
bons | -0.060 |
bars | -0.054 |
shirts | -0.051 |
cans | -0.050 |
growth | -0.050 |
Token | on |
Token ID | 319 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
bons | +0.061 |
growth | +0.055 |
talk | +0.055 |
ests | +0.052 |
bars | +0.051 |
Neg logit contributions | |
glad | -0.042 |
pleased | -0.039 |
okay | -0.035 |
lucky | -0.035 |
impressed | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
ests | +0.267 |
talk | +0.262 |
bons | +0.256 |
ales | +0.253 |
orns | +0.240 |
Neg logit contributions | |
glad | -0.079 |
happy | -0.071 |
lucky | -0.068 |
sad | -0.065 |
okay | -0.060 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
bons | +0.094 |
talk | +0.091 |
growth | +0.084 |
ales | +0.083 |
ests | +0.082 |
Neg logit contributions | |
glad | -0.105 |
lucky | -0.098 |
pleased | -0.093 |
okay | -0.092 |
sorry | -0.089 |
Token | box |
Token ID | 3091 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
bons | +0.256 |
ales | +0.232 |
growth | +0.228 |
talk | +0.226 |
aled | +0.213 |
Neg logit contributions | |
sorry | -0.158 |
okay | -0.153 |
glad | -0.151 |
sad | -0.148 |
lucky | -0.147 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
bons | +0.106 |
growth | +0.098 |
talk | +0.097 |
ests | +0.090 |
ales | +0.088 |
Neg logit contributions | |
glad | -0.081 |
pleased | -0.075 |
okay | -0.072 |
sorry | -0.069 |
impressed | -0.069 |
Token | She |
Token ID | 1375 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
bons | +0.088 |
talk | +0.078 |
ales | +0.075 |
growth | +0.073 |
stations | +0.071 |
Neg logit contributions | |
glad | -0.089 |
pleased | -0.080 |
sorry | -0.078 |
lucky | -0.077 |
okay | -0.076 |
Token | thinks |
Token ID | 6834 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
bons | +0.060 |
talk | +0.053 |
growth | +0.051 |
ales | +0.050 |
ests | +0.049 |
Neg logit contributions | |
glad | -0.073 |
pleased | -0.068 |
impressed | -0.064 |
lucky | -0.064 |
sorry | -0.063 |
Token | : |
Token ID | 25 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
glad | +0.021 |
pleased | +0.019 |
impressed | +0.017 |
lucky | +0.017 |
okay | +0.017 |
Neg logit contributions | |
bons | -0.025 |
growth | -0.022 |
bars | -0.022 |
talk | -0.022 |
shirts | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
talk | +0.197 |
bons | +0.189 |
ests | +0.189 |
ales | +0.182 |
growth | +0.173 |
Neg logit contributions | |
glad | -0.104 |
okay | -0.094 |
happy | -0.093 |
lucky | -0.092 |
pleased | -0.089 |
Token | circle |
Token ID | 9197 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
bons | +0.041 |
talk | +0.038 |
growth | +0.035 |
ests | +0.032 |
ales | +0.031 |
Neg logit contributions | |
glad | -0.094 |
pleased | -0.090 |
lucky | -0.087 |
sorry | -0.086 |
okay | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
bons | +0.162 |
growth | +0.152 |
talk | +0.150 |
ales | +0.138 |
ests | +0.136 |
Neg logit contributions | |
glad | -0.133 |
pleased | -0.127 |
okay | -0.115 |
lucky | -0.113 |
impressed | -0.112 |
Token | stuck |
Token ID | 7819 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
bons | +0.105 |
talk | +0.096 |
ests | +0.094 |
ales | +0.092 |
growth | +0.092 |
Neg logit contributions | |
glad | -0.122 |
pleased | -0.117 |
impressed | -0.110 |
sorry | -0.108 |
okay | -0.108 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
bons | +0.104 |
talk | +0.101 |
growth | +0.095 |
ests | +0.094 |
ales | +0.091 |
Neg logit contributions | |
glad | -0.077 |
pleased | -0.070 |
okay | -0.068 |
sorry | -0.066 |
lucky | -0.065 |
Token | pet |
Token ID | 4273 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
bons | +0.041 |
talk | +0.036 |
growth | +0.035 |
ests | +0.031 |
ales | +0.031 |
Neg logit contributions | |
glad | -0.085 |
pleased | -0.081 |
lucky | -0.078 |
sorry | -0.077 |
okay | -0.077 |
Token | als |
Token ID | 874 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
bons | +0.131 |
talk | +0.125 |
growth | +0.111 |
ests | +0.108 |
ales | +0.098 |
Neg logit contributions | |
glad | -0.267 |
lucky | -0.259 |
okay | -0.255 |
sorry | -0.255 |
pleased | -0.248 |
Token | on |
Token ID | 319 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
bons | +0.087 |
talk | +0.079 |
growth | +0.078 |
ests | +0.075 |
shirts | +0.072 |
Neg logit contributions | |
glad | -0.069 |
pleased | -0.065 |
okay | -0.060 |
lucky | -0.058 |
sorry | -0.058 |
Token | it |
Token ID | 340 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
talk | +0.210 |
ests | +0.202 |
bons | +0.202 |
ales | +0.192 |
orns | +0.181 |
Neg logit contributions | |
glad | -0.102 |
okay | -0.094 |
happy | -0.089 |
lucky | -0.089 |
sorry | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
bons | +0.111 |
talk | +0.102 |
growth | +0.102 |
ests | +0.098 |
ales | +0.094 |
Neg logit contributions | |
glad | -0.086 |
pleased | -0.080 |
okay | -0.077 |
lucky | -0.074 |
sorry | -0.073 |
Token | She |
Token ID | 1375 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
bons | +0.017 |
talk | +0.015 |
growth | +0.015 |
bars | +0.014 |
ests | +0.014 |
Neg logit contributions | |
glad | -0.019 |
pleased | -0.017 |
impressed | -0.016 |
okay | -0.016 |
lucky | -0.016 |
Token | he |
Token ID | 339 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
glad | +0.020 |
pleased | +0.019 |
impressed | +0.017 |
lucky | +0.016 |
sorry | +0.016 |
Neg logit contributions | |
bons | -0.021 |
growth | -0.019 |
bars | -0.018 |
talk | -0.018 |
shirts | -0.017 |
Token | put |
Token ID | 1234 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
bons | +0.079 |
growth | +0.070 |
talk | +0.068 |
ales | +0.065 |
ests | +0.065 |
Neg logit contributions | |
glad | -0.065 |
pleased | -0.062 |
impressed | -0.058 |
okay | -0.057 |
lucky | -0.057 |
Token | it |
Token ID | 340 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
bons | +0.088 |
talk | +0.084 |
growth | +0.084 |
bars | +0.079 |
ales | +0.078 |
Neg logit contributions | |
glad | -0.056 |
pleased | -0.049 |
lucky | -0.046 |
sorry | -0.045 |
okay | -0.045 |
Token | on |
Token ID | 319 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
bons | +0.030 |
talk | +0.027 |
growth | +0.027 |
bars | +0.026 |
ests | +0.025 |
Neg logit contributions | |
glad | -0.022 |
pleased | -0.021 |
okay | -0.019 |
impressed | -0.019 |
lucky | -0.018 |
Token | his |
Token ID | 465 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
bons | +0.151 |
talk | +0.147 |
als | +0.143 |
ests | +0.143 |
ales | +0.141 |
Neg logit contributions | |
glad | -0.094 |
pleased | -0.085 |
okay | -0.083 |
lucky | -0.079 |
happy | -0.079 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
bons | +0.070 |
talk | +0.069 |
growth | +0.066 |
ales | +0.061 |
als | +0.059 |
Neg logit contributions | |
glad | -0.086 |
pleased | -0.079 |
lucky | -0.079 |
okay | -0.076 |
sorry | -0.076 |
Token | shelf |
Token ID | 18316 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
growth | +0.221 |
bons | +0.220 |
talk | +0.206 |
ales | +0.196 |
aled | +0.193 |
Neg logit contributions | |
glad | -0.171 |
sorry | -0.170 |
sad | -0.167 |
lucky | -0.165 |
okay | -0.164 |
Token | . |
Token ID | 13 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
bons | +0.123 |
growth | +0.113 |
talk | +0.106 |
ests | +0.100 |
bars | +0.099 |
Neg logit contributions | |
glad | -0.098 |
pleased | -0.090 |
okay | -0.084 |
impressed | -0.082 |
lucky | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | +9.08e-01 |
Pos logit contributions | |
bons | +0.034 |
talk | +0.030 |
growth | +0.030 |
bars | +0.029 |
ales | +0.028 |
Neg logit contributions | |
glad | -0.026 |
pleased | -0.024 |
okay | -0.022 |
impressed | -0.022 |
lucky | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
glad | +0.051 |
pleased | +0.049 |
sorry | +0.045 |
okay | +0.044 |
impressed | +0.044 |
Neg logit contributions | |
bons | -0.048 |
talk | -0.045 |
growth | -0.043 |
bars | -0.042 |
ests | -0.041 |
Token | Later |
Token ID | 18602 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
bons | +0.012 |
talk | +0.010 |
growth | +0.010 |
bars | +0.010 |
ests | +0.009 |
Neg logit contributions | |
glad | -0.014 |
pleased | -0.013 |
impressed | -0.012 |
okay | -0.012 |
sorry | -0.012 |
Token | would |
Token ID | 561 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
bons | +0.076 |
growth | +0.067 |
talk | +0.066 |
ales | +0.064 |
bars | +0.062 |
Neg logit contributions | |
glad | -0.079 |
pleased | -0.074 |
impressed | -0.070 |
lucky | -0.069 |
okay | -0.067 |
Token | take |
Token ID | 1011 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
bons | +0.095 |
ests | +0.085 |
talk | +0.085 |
growth | +0.084 |
orns | +0.081 |
Neg logit contributions | |
glad | -0.139 |
pleased | -0.131 |
impressed | -0.125 |
sorry | -0.123 |
lucky | -0.123 |
Token | out |
Token ID | 503 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
bons | +0.102 |
talk | +0.098 |
growth | +0.096 |
ales | +0.094 |
ests | +0.091 |
Neg logit contributions | |
glad | -0.096 |
pleased | -0.088 |
lucky | -0.088 |
sorry | -0.083 |
happy | -0.081 |
Token | his |
Token ID | 465 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
bons | +0.123 |
talk | +0.117 |
growth | +0.112 |
ales | +0.108 |
bars | +0.106 |
Neg logit contributions | |
glad | -0.095 |
pleased | -0.085 |
lucky | -0.082 |
sorry | -0.079 |
impressed | -0.077 |
Token | colored |
Token ID | 16396 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
bons | +0.056 |
talk | +0.053 |
growth | +0.049 |
ales | +0.045 |
ests | +0.043 |
Neg logit contributions | |
glad | -0.104 |
pleased | -0.097 |
lucky | -0.096 |
sorry | -0.092 |
okay | -0.091 |
Token | chalk |
Token ID | 30860 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
talk | +0.167 |
growth | +0.165 |
ales | +0.151 |
bons | +0.150 |
keeping | +0.142 |
Neg logit contributions | |
glad | -0.192 |
pleased | -0.192 |
lucky | -0.191 |
happy | -0.190 |
sorry | -0.183 |
Token | and |
Token ID | 290 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
growth | +0.286 |
bons | +0.283 |
stations | +0.260 |
ales | +0.260 |
Mine | +0.257 |
Neg logit contributions | |
glad | -0.113 |
happy | -0.111 |
lucky | -0.110 |
pleased | -0.109 |
okay | -0.104 |
Token | proudly |
Token ID | 23017 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
bons | +0.078 |
ests | +0.071 |
talk | +0.070 |
growth | +0.067 |
ales | +0.065 |
Neg logit contributions | |
glad | -0.121 |
pleased | -0.114 |
sorry | -0.108 |
lucky | -0.108 |
impressed | -0.108 |
Token | draw |
Token ID | 3197 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
bons | +0.049 |
growth | +0.043 |
talk | +0.042 |
bars | +0.039 |
ests | +0.039 |
Neg logit contributions | |
glad | -0.072 |
pleased | -0.069 |
impressed | -0.064 |
lucky | -0.063 |
sorry | -0.063 |
Token | pictures |
Token ID | 5986 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
bons | +0.084 |
talk | +0.081 |
growth | +0.080 |
ales | +0.075 |
bars | +0.070 |
Neg logit contributions | |
glad | -0.100 |
pleased | -0.097 |
impressed | -0.092 |
lucky | -0.091 |
okay | -0.089 |
Token | on |
Token ID | 319 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
bons | +0.119 |
growth | +0.116 |
talk | +0.112 |
ales | +0.105 |
bars | +0.101 |
Neg logit contributions | |
glad | -0.083 |
pleased | -0.078 |
impressed | -0.071 |
okay | -0.070 |
lucky | -0.069 |
Token | draw |
Token ID | 3197 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
bons | +0.072 |
talk | +0.066 |
growth | +0.061 |
bars | +0.060 |
ests | +0.060 |
Neg logit contributions | |
glad | -0.086 |
pleased | -0.079 |
sorry | -0.076 |
lucky | -0.076 |
okay | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
bons | +0.098 |
growth | +0.097 |
talk | +0.096 |
ales | +0.089 |
bars | +0.082 |
Neg logit contributions | |
glad | -0.092 |
pleased | -0.088 |
impressed | -0.084 |
lucky | -0.083 |
happy | -0.081 |
Token | write |
Token ID | 3551 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
bons | +0.042 |
talk | +0.037 |
growth | +0.036 |
ests | +0.034 |
ales | +0.033 |
Neg logit contributions | |
glad | -0.073 |
pleased | -0.069 |
impressed | -0.065 |
lucky | -0.065 |
sorry | -0.064 |
Token | on |
Token ID | 319 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
bons | +0.019 |
talk | +0.017 |
growth | +0.017 |
bars | +0.016 |
ests | +0.015 |
Neg logit contributions | |
glad | -0.019 |
pleased | -0.017 |
impressed | -0.016 |
okay | -0.016 |
lucky | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
talk | +0.133 |
bons | +0.124 |
ests | +0.120 |
growth | +0.117 |
ales | +0.115 |
Neg logit contributions | |
glad | -0.127 |
lucky | -0.120 |
pleased | -0.116 |
happy | -0.115 |
impressed | -0.112 |
Token | black |
Token ID | 2042 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
bons | +0.075 |
talk | +0.071 |
growth | +0.069 |
ests | +0.063 |
ales | +0.062 |
Neg logit contributions | |
glad | -0.118 |
pleased | -0.112 |
lucky | -0.112 |
impressed | -0.106 |
sorry | -0.105 |
Token | board |
Token ID | 3526 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
bons | +0.154 |
talk | +0.139 |
ales | +0.132 |
aled | +0.122 |
growth | +0.119 |
Neg logit contributions | |
lucky | -0.217 |
glad | -0.209 |
sad | -0.207 |
happy | -0.207 |
pleased | -0.203 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
bons | +0.132 |
growth | +0.126 |
talk | +0.119 |
ales | +0.113 |
ests | +0.112 |
Neg logit contributions | |
glad | -0.096 |
pleased | -0.090 |
sorry | -0.083 |
impressed | -0.083 |
lucky | -0.082 |
Token | One |
Token ID | 1881 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
bons | +0.069 |
talk | +0.064 |
growth | +0.063 |
bars | +0.058 |
ales | +0.058 |
Neg logit contributions | |
glad | -0.072 |
pleased | -0.064 |
okay | -0.064 |
sorry | -0.061 |
lucky | -0.061 |
Token | day |
Token ID | 1110 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
bons | +0.030 |
talk | +0.027 |
growth | +0.027 |
bars | +0.026 |
ests | +0.025 |
Neg logit contributions | |
glad | -0.022 |
pleased | -0.020 |
lucky | -0.018 |
okay | -0.018 |
impressed | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
glad | +0.083 |
pleased | +0.083 |
impressed | +0.078 |
lucky | +0.069 |
okay | +0.068 |
Neg logit contributions | |
bons | -0.067 |
bars | -0.061 |
ted | -0.059 |
growth | -0.059 |
shirts | -0.059 |
Token | He |
Token ID | 679 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
bons | +0.081 |
talk | +0.069 |
ales | +0.068 |
growth | +0.068 |
bars | +0.064 |
Neg logit contributions | |
glad | -0.079 |
sorry | -0.071 |
pleased | -0.070 |
okay | -0.067 |
lucky | -0.065 |
Token | tri |
Token ID | 1333 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
bons | +0.098 |
growth | +0.087 |
talk | +0.084 |
ales | +0.082 |
ests | +0.081 |
Neg logit contributions | |
glad | -0.067 |
pleased | -0.061 |
impressed | -0.057 |
lucky | -0.056 |
sorry | -0.055 |
Token | pped |
Token ID | 1496 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
bons | +0.055 |
talk | +0.046 |
growth | +0.046 |
ales | +0.043 |
bars | +0.043 |
Neg logit contributions | |
glad | -0.066 |
pleased | -0.061 |
okay | -0.057 |
impressed | -0.057 |
sorry | -0.057 |
Token | on |
Token ID | 319 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
bons | +0.184 |
growth | +0.168 |
ales | +0.164 |
talk | +0.164 |
ests | +0.161 |
Neg logit contributions | |
glad | -0.115 |
sad | -0.099 |
sorry | -0.098 |
okay | -0.096 |
happy | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
talk | +0.224 |
bons | +0.222 |
ests | +0.212 |
ales | +0.211 |
board | +0.202 |
Neg logit contributions | |
glad | -0.100 |
sorry | -0.092 |
okay | -0.086 |
lucky | -0.083 |
sad | -0.082 |
Token | rug |
Token ID | 14477 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
bons | +0.056 |
talk | +0.050 |
growth | +0.049 |
ales | +0.045 |
ests | +0.043 |
Neg logit contributions | |
glad | -0.096 |
pleased | -0.090 |
sorry | -0.089 |
lucky | -0.086 |
okay | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
bons | +0.303 |
growth | +0.267 |
ests | +0.262 |
talk | +0.260 |
ales | +0.255 |
Neg logit contributions | |
glad | -0.154 |
sorry | -0.154 |
sad | -0.152 |
okay | -0.146 |
pleased | -0.140 |
Token | fell |
Token ID | 3214 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
bons | +0.097 |
talk | +0.086 |
ales | +0.084 |
growth | +0.084 |
ests | +0.082 |
Neg logit contributions | |
glad | -0.066 |
pleased | -0.057 |
sorry | -0.057 |
lucky | -0.054 |
okay | -0.053 |
Token | down |
Token ID | 866 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
bons | +0.046 |
growth | +0.041 |
talk | +0.041 |
bars | +0.039 |
ales | +0.038 |
Neg logit contributions | |
glad | -0.039 |
pleased | -0.036 |
okay | -0.033 |
sorry | -0.033 |
lucky | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -9.78e-01 |
Pos logit contributions | |
bons | +0.107 |
talk | +0.094 |
ales | +0.094 |
growth | +0.092 |
shirts | +0.090 |
Neg logit contributions | |
glad | -0.076 |
pleased | -0.069 |
sorry | -0.066 |
lucky | -0.063 |
okay | -0.063 |
Token | He |
Token ID | 679 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
bons | +0.054 |
talk | +0.047 |
growth | +0.045 |
ales | +0.045 |
bars | +0.044 |
Neg logit contributions | |
glad | -0.059 |
pleased | -0.054 |
sorry | -0.052 |
okay | -0.050 |
lucky | -0.050 |
Token | in |
Token ID | 287 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
glad | +0.008 |
pleased | +0.008 |
impressed | +0.007 |
sorry | +0.007 |
okay | +0.007 |
Neg logit contributions | |
bons | -0.009 |
talk | -0.008 |
growth | -0.008 |
bars | -0.007 |
ales | -0.007 |
Token | their |
Token ID | 511 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
bons | +0.025 |
talk | +0.022 |
growth | +0.022 |
bars | +0.021 |
ests | +0.020 |
Neg logit contributions | |
glad | -0.024 |
pleased | -0.022 |
impressed | -0.020 |
okay | -0.020 |
lucky | -0.020 |
Token | world |
Token ID | 995 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
bons | +0.012 |
talk | +0.010 |
growth | +0.010 |
bars | +0.010 |
shirts | +0.009 |
Neg logit contributions | |
glad | -0.015 |
pleased | -0.014 |
impressed | -0.013 |
okay | -0.013 |
sorry | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
bons | +0.018 |
talk | +0.016 |
growth | +0.016 |
bars | +0.015 |
ests | +0.014 |
Neg logit contributions | |
glad | -0.019 |
pleased | -0.018 |
impressed | -0.017 |
okay | -0.017 |
lucky | -0.016 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
bons | +0.002 |
talk | +0.001 |
growth | +0.001 |
bars | +0.001 |
ests | +0.001 |
Neg logit contributions | |
glad | -0.001 |
pleased | -0.001 |
impressed | -0.001 |
okay | -0.001 |
sorry | -0.001 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
bons | +0.003 |
talk | +0.003 |
growth | +0.003 |
bars | +0.003 |
ests | +0.003 |
Neg logit contributions | |
glad | -0.003 |
pleased | -0.003 |
impressed | -0.002 |
okay | -0.002 |
sorry | -0.002 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
pleased | +0.163 |
glad | +0.157 |
sorry | +0.148 |
impressed | +0.144 |
okay | +0.141 |
Neg logit contributions | |
ests | -0.181 |
boxes | -0.180 |
bons | -0.180 |
talk | -0.179 |
ales | -0.178 |
Token | a |
Token ID | 257 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
glad | +0.035 |
pleased | +0.033 |
impressed | +0.030 |
sorry | +0.030 |
lucky | +0.029 |
Neg logit contributions | |
bons | -0.032 |
growth | -0.029 |
talk | -0.027 |
bars | -0.027 |
shirts | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
glad | +0.092 |
pleased | +0.089 |
impressed | +0.083 |
sorry | +0.083 |
lucky | +0.077 |
Neg logit contributions | |
bons | -0.087 |
growth | -0.079 |
talk | -0.077 |
bars | -0.076 |
shirts | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
pleased | +0.241 |
glad | +0.226 |
impressed | +0.220 |
thankful | +0.207 |
grateful | +0.206 |
Neg logit contributions | |
bars | -0.093 |
room | -0.092 |
bons | -0.091 |
cans | -0.085 |
growth | -0.084 |
Token | there |
Token ID | 612 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
pleased | +0.145 |
glad | +0.132 |
impressed | +0.131 |
thankful | +0.112 |
blessed | +0.110 |
Neg logit contributions | |
bars | -0.124 |
bons | -0.120 |
growth | -0.119 |
cans | -0.116 |
tape | -0.115 |
Token | wet |
Token ID | 9583 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
glad | +0.058 |
pleased | +0.055 |
impressed | +0.050 |
lucky | +0.047 |
sorry | +0.046 |
Neg logit contributions | |
bons | -0.064 |
cans | -0.055 |
boxes | -0.054 |
growth | -0.054 |
bars | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
bons | +0.044 |
talk | +0.039 |
growth | +0.039 |
bars | +0.037 |
ests | +0.035 |
Neg logit contributions | |
glad | -0.047 |
pleased | -0.044 |
okay | -0.042 |
lucky | -0.041 |
sorry | -0.041 |
Token | Max |
Token ID | 5436 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
bons | +0.028 |
talk | +0.025 |
growth | +0.024 |
bars | +0.023 |
ests | +0.022 |
Neg logit contributions | |
glad | -0.036 |
pleased | -0.034 |
sorry | -0.031 |
okay | -0.031 |
impressed | -0.031 |
Token | 's |
Token ID | 338 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
glad | +0.059 |
pleased | +0.055 |
sorry | +0.052 |
okay | +0.052 |
lucky | +0.050 |
Neg logit contributions | |
bons | -0.043 |
bars | -0.038 |
talk | -0.038 |
shirts | -0.037 |
growth | -0.037 |
Token | dad |
Token ID | 9955 |
Feature activation | +9.65e-01 |
Pos logit contributions | |
glad | +0.025 |
pleased | +0.024 |
impressed | +0.021 |
okay | +0.021 |
sorry | +0.021 |
Neg logit contributions | |
bons | -0.027 |
growth | -0.025 |
talk | -0.024 |
shirts | -0.024 |
bars | -0.024 |
Token | said |
Token ID | 531 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
glad | +0.058 |
pleased | +0.054 |
okay | +0.050 |
lucky | +0.050 |
sorry | +0.050 |
Neg logit contributions | |
bons | -0.045 |
talk | -0.040 |
bars | -0.040 |
shirts | -0.038 |
growth | -0.037 |
Token | " |
Token ID | 366 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
glad | +0.169 |
pleased | +0.165 |
impressed | +0.155 |
shadow | +0.153 |
lucky | +0.145 |
Neg logit contributions | |
cans | -0.088 |
ted | -0.083 |
bars | -0.082 |
als | -0.082 |
shirts | -0.081 |
Token | We |
Token ID | 1135 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
bons | +0.082 |
growth | +0.074 |
bars | +0.072 |
talk | +0.069 |
shirts | +0.067 |
Neg logit contributions | |
glad | -0.087 |
pleased | -0.085 |
impressed | -0.077 |
okay | -0.077 |
lucky | -0.074 |
Token | need |
Token ID | 761 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
glad | +0.012 |
pleased | +0.011 |
impressed | +0.010 |
lucky | +0.010 |
sorry | +0.010 |
Neg logit contributions | |
bons | -0.010 |
talk | -0.009 |
growth | -0.009 |
bars | -0.009 |
ales | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
bons | +0.001 |
talk | +0.001 |
growth | +0.001 |
bars | +0.001 |
shirts | +0.001 |
Neg logit contributions | |
glad | -0.001 |
pleased | -0.001 |
impressed | -0.001 |
lucky | -0.001 |
sorry | -0.001 |
Token | work |
Token ID | 670 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
bons | +0.043 |
talk | +0.039 |
growth | +0.038 |
bars | +0.036 |
ests | +0.036 |
Neg logit contributions | |
glad | -0.052 |
pleased | -0.048 |
okay | -0.046 |
sorry | -0.045 |
lucky | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
glad | +0.041 |
pleased | +0.038 |
impressed | +0.035 |
sorry | +0.035 |
okay | +0.035 |
Neg logit contributions | |
bons | -0.039 |
growth | -0.034 |
talk | -0.034 |
bars | -0.033 |
shirts | -0.032 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
bons | +0.011 |
talk | +0.010 |
growth | +0.010 |
bars | +0.010 |
ests | +0.009 |
Neg logit contributions | |
glad | -0.009 |
pleased | -0.008 |
impressed | -0.008 |
okay | -0.008 |
lucky | -0.007 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
bons | +0.024 |
talk | +0.022 |
growth | +0.021 |
bars | +0.021 |
ests | +0.020 |
Neg logit contributions | |
glad | -0.019 |
pleased | -0.017 |
impressed | -0.015 |
okay | -0.015 |
lucky | -0.015 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
pleased | +0.165 |
glad | +0.156 |
sorry | +0.151 |
okay | +0.149 |
impressed | +0.147 |
Neg logit contributions | |
boxes | -0.179 |
els | -0.176 |
shirts | -0.176 |
board | -0.174 |
talk | -0.173 |
Token | a |
Token ID | 257 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
glad | +0.036 |
pleased | +0.034 |
impressed | +0.031 |
sorry | +0.031 |
lucky | +0.030 |
Neg logit contributions | |
bons | -0.032 |
growth | -0.029 |
talk | -0.027 |
bars | -0.027 |
shirts | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
glad | +0.108 |
pleased | +0.106 |
impressed | +0.100 |
sorry | +0.098 |
lucky | +0.091 |
Neg logit contributions | |
bons | -0.094 |
growth | -0.085 |
talk | -0.082 |
bars | -0.081 |
shirts | -0.078 |
Token | , |
Token ID | 11 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
pleased | +0.267 |
glad | +0.251 |
impressed | +0.244 |
thankful | +0.233 |
grateful | +0.231 |
Neg logit contributions | |
room | -0.094 |
bars | -0.083 |
cans | -0.080 |
bons | -0.080 |
board | -0.077 |
Token | there |
Token ID | 612 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
pleased | +0.165 |
glad | +0.152 |
impressed | +0.149 |
sorry | +0.133 |
blessed | +0.132 |
Neg logit contributions | |
bars | -0.123 |
growth | -0.122 |
room | -0.121 |
bons | -0.120 |
tape | -0.119 |
Token | was |
Token ID | 373 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
glad | +0.028 |
pleased | +0.027 |
impressed | +0.025 |
sorry | +0.025 |
okay | +0.024 |
Neg logit contributions | |
bons | -0.021 |
growth | -0.018 |
bars | -0.018 |
talk | -0.018 |
ests | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
bons | +0.022 |
talk | +0.020 |
growth | +0.020 |
bars | +0.019 |
ests | +0.018 |
Neg logit contributions | |
glad | -0.019 |
pleased | -0.018 |
impressed | -0.016 |
okay | -0.016 |
lucky | -0.016 |
Token | human |
Token ID | 1692 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
glad | +0.115 |
pleased | +0.112 |
okay | +0.104 |
shadow | +0.101 |
ok | +0.101 |
Neg logit contributions | |
bons | -0.116 |
bars | -0.112 |
shirts | -0.112 |
paper | -0.110 |
cash | -0.108 |
Token | - |
Token ID | 12 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
glad | +0.074 |
pleased | +0.069 |
sorry | +0.064 |
impressed | +0.063 |
okay | +0.063 |
Neg logit contributions | |
bons | -0.060 |
talk | -0.055 |
bars | -0.054 |
growth | -0.051 |
als | -0.050 |
Token | Ch |
Token ID | 1925 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
bons | +0.004 |
talk | +0.004 |
growth | +0.003 |
bars | +0.003 |
ests | +0.003 |
Neg logit contributions | |
glad | -0.008 |
pleased | -0.008 |
impressed | -0.007 |
okay | -0.007 |
sorry | -0.007 |
Token | oo |
Token ID | 2238 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
bons | +0.009 |
talk | +0.008 |
growth | +0.008 |
bars | +0.008 |
ests | +0.007 |
Neg logit contributions | |
glad | -0.006 |
pleased | -0.005 |
impressed | -0.005 |
okay | -0.005 |
lucky | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
glad | +0.062 |
pleased | +0.058 |
sorry | +0.054 |
impressed | +0.054 |
okay | +0.053 |
Neg logit contributions | |
bons | -0.051 |
talk | -0.046 |
bars | -0.045 |
growth | -0.043 |
als | -0.042 |
Token | feeling |
Token ID | 4203 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
pleased | +0.141 |
glad | +0.128 |
reetings | +0.125 |
impressed | +0.123 |
lad | +0.117 |
Neg logit contributions | |
bars | -0.118 |
bons | -0.114 |
pedals | -0.103 |
stick | -0.100 |
shelves | -0.099 |
Token | very |
Token ID | 845 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
pleased | +0.086 |
glad | +0.084 |
impressed | +0.077 |
reetings | +0.072 |
shadow | +0.072 |
Neg logit contributions | |
bons | -0.124 |
bars | -0.110 |
cans | -0.110 |
shelves | -0.108 |
fabric | -0.107 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
pleased | +0.117 |
impressed | +0.113 |
shadow | +0.112 |
reetings | +0.110 |
glad | +0.109 |
Neg logit contributions | |
shelves | -0.153 |
bons | -0.151 |
grass | -0.151 |
rocks | -0.151 |
leaves | -0.149 |
Token | because |
Token ID | 780 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
bons | +0.016 |
talk | +0.014 |
growth | +0.014 |
bars | +0.013 |
ests | +0.013 |
Neg logit contributions | |
glad | -0.017 |
pleased | -0.016 |
impressed | -0.015 |
okay | -0.015 |
lucky | -0.015 |
Token | he |
Token ID | 339 |
Feature activation | +9.35e-01 |
Pos logit contributions | |
glad | +0.010 |
pleased | +0.010 |
impressed | +0.009 |
sorry | +0.009 |
okay | +0.009 |
Neg logit contributions | |
bons | -0.012 |
growth | -0.011 |
talk | -0.011 |
bars | -0.010 |
ests | -0.010 |
Token | had |
Token ID | 550 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
glad | +0.056 |
pleased | +0.054 |
sorry | +0.052 |
okay | +0.051 |
impressed | +0.048 |
Neg logit contributions | |
bons | -0.045 |
talk | -0.041 |
growth | -0.039 |
bars | -0.039 |
als | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
glad | +0.060 |
pleased | +0.059 |
sorry | +0.055 |
impressed | +0.053 |
okay | +0.051 |
Neg logit contributions | |
bons | -0.059 |
talk | -0.050 |
bars | -0.050 |
growth | -0.050 |
shirts | -0.047 |
Token | Oh |
Token ID | 5812 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
bons | +0.022 |
talk | +0.019 |
growth | +0.019 |
bars | +0.018 |
ests | +0.017 |
Neg logit contributions | |
glad | -0.027 |
pleased | -0.026 |
impressed | -0.024 |
okay | -0.024 |
lucky | -0.024 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
pleased | +0.084 |
glad | +0.083 |
impressed | +0.079 |
lucky | +0.067 |
grateful | +0.065 |
Neg logit contributions | |
bons | -0.081 |
bars | -0.075 |
growth | -0.070 |
als | -0.070 |
cans | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
glad | +0.066 |
pleased | +0.064 |
impressed | +0.060 |
lucky | +0.054 |
sorry | +0.054 |
Neg logit contributions | |
bons | -0.073 |
bars | -0.066 |
als | -0.064 |
talk | -0.063 |
growth | -0.061 |
Token | I |
Token ID | 314 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
glad | +0.076 |
pleased | +0.075 |
impressed | +0.068 |
lucky | +0.061 |
sorry | +0.060 |
Neg logit contributions | |
bons | -0.081 |
bars | -0.073 |
shirts | -0.071 |
cans | -0.071 |
growth | -0.071 |
Token | 'm |
Token ID | 1101 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
glad | +0.063 |
pleased | +0.060 |
impressed | +0.055 |
sorry | +0.053 |
lucky | +0.053 |
Neg logit contributions | |
bons | -0.060 |
talk | -0.058 |
growth | -0.055 |
bars | -0.054 |
cans | -0.050 |
Token | so |
Token ID | 523 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
pleased | +0.081 |
reetings | +0.074 |
glad | +0.066 |
impressed | +0.066 |
rav | +0.064 |
Neg logit contributions | |
bons | -0.157 |
bars | -0.151 |
cans | -0.150 |
boxes | -0.142 |
glue | -0.141 |
Token | proud |
Token ID | 6613 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
shadow | +0.102 |
reetings | +0.093 |
pleased | +0.088 |
okay | +0.076 |
ok | +0.076 |
Neg logit contributions | |
cans | -0.180 |
leaves | -0.170 |
papers | -0.165 |
paint | -0.164 |
glue | -0.163 |
Token | of |
Token ID | 286 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
glad | +0.047 |
pleased | +0.045 |
impressed | +0.041 |
sorry | +0.041 |
lucky | +0.040 |
Neg logit contributions | |
bons | -0.038 |
talk | -0.032 |
bars | -0.031 |
growth | -0.031 |
shirts | -0.030 |
Token | you |
Token ID | 345 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
pleased | +0.048 |
glad | +0.048 |
impressed | +0.043 |
lucky | +0.040 |
okay | +0.040 |
Neg logit contributions | |
bons | -0.050 |
growth | -0.042 |
bars | -0.040 |
shirts | -0.040 |
talk | -0.039 |
Token | for |
Token ID | 329 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
glad | +0.101 |
pleased | +0.098 |
impressed | +0.091 |
lucky | +0.086 |
sorry | +0.085 |
Neg logit contributions | |
bons | -0.078 |
bars | -0.067 |
shirts | -0.065 |
talk | -0.063 |
als | -0.063 |
Token | making |
Token ID | 1642 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
glad | +0.048 |
pleased | +0.047 |
impressed | +0.041 |
okay | +0.040 |
lucky | +0.040 |
Neg logit contributions | |
bons | -0.063 |
growth | -0.054 |
shirts | -0.053 |
bars | -0.053 |
cans | -0.052 |
Token | They |
Token ID | 1119 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
glad | +0.002 |
pleased | +0.002 |
impressed | +0.002 |
lucky | +0.002 |
okay | +0.002 |
Neg logit contributions | |
bons | -0.002 |
talk | -0.001 |
growth | -0.001 |
bars | -0.001 |
ests | -0.001 |
Token | used |
Token ID | 973 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
bons | +0.064 |
talk | +0.056 |
growth | +0.056 |
ests | +0.054 |
bars | +0.053 |
Neg logit contributions | |
glad | -0.067 |
pleased | -0.063 |
impressed | -0.058 |
lucky | -0.057 |
sorry | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
bons | +0.068 |
talk | +0.063 |
growth | +0.060 |
bars | +0.058 |
ests | +0.058 |
Neg logit contributions | |
glad | -0.063 |
pleased | -0.059 |
impressed | -0.054 |
lucky | -0.053 |
okay | -0.053 |
Token | stick |
Token ID | 4859 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
glad | +0.012 |
pleased | +0.012 |
impressed | +0.011 |
okay | +0.011 |
sorry | +0.011 |
Neg logit contributions | |
bons | -0.010 |
talk | -0.008 |
growth | -0.008 |
bars | -0.008 |
shirts | -0.008 |
Token | as |
Token ID | 355 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
bons | +0.094 |
growth | +0.086 |
talk | +0.086 |
ests | +0.080 |
bars | +0.080 |
Neg logit contributions | |
glad | -0.061 |
pleased | -0.056 |
okay | -0.050 |
impressed | -0.050 |
lucky | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
bons | +0.001 |
growth | +0.001 |
talk | +0.001 |
bars | +0.001 |
shirts | +0.001 |
Neg logit contributions | |
glad | -0.001 |
pleased | -0.001 |
impressed | -0.000 |
okay | -0.000 |
sorry | -0.000 |
Token | flag |
Token ID | 6056 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
glad | +0.050 |
pleased | +0.045 |
impressed | +0.043 |
okay | +0.042 |
sorry | +0.042 |
Neg logit contributions | |
bons | -0.039 |
growth | -0.034 |
bars | -0.033 |
talk | -0.033 |
shirts | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
bons | +0.047 |
talk | +0.042 |
growth | +0.042 |
bars | +0.039 |
ests | +0.039 |
Neg logit contributions | |
glad | -0.040 |
pleased | -0.038 |
okay | -0.034 |
impressed | -0.034 |
lucky | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
bons | +0.061 |
talk | +0.055 |
growth | +0.052 |
bars | +0.051 |
ales | +0.050 |
Neg logit contributions | |
glad | -0.073 |
pleased | -0.067 |
okay | -0.064 |
sorry | -0.063 |
impressed | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
glad | +0.026 |
pleased | +0.025 |
sorry | +0.023 |
okay | +0.023 |
lucky | +0.022 |
Neg logit contributions | |
bons | -0.028 |
talk | -0.025 |
growth | -0.025 |
bars | -0.024 |
ests | -0.024 |
Token | They |
Token ID | 2990 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
bons | +0.009 |
talk | +0.008 |
growth | +0.008 |
bars | +0.008 |
ests | +0.007 |
Neg logit contributions | |
glad | -0.010 |
pleased | -0.009 |
okay | -0.008 |
impressed | -0.008 |
lucky | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
glad | +0.116 |
pleased | +0.114 |
impressed | +0.107 |
sorry | +0.103 |
lucky | +0.101 |
Neg logit contributions | |
bons | -0.082 |
bars | -0.079 |
cans | -0.069 |
talk | -0.068 |
boxes | -0.067 |
Token | help |
Token ID | 1037 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
bons | +0.050 |
talk | +0.045 |
growth | +0.043 |
ests | +0.042 |
shirts | +0.042 |
Neg logit contributions | |
glad | -0.051 |
pleased | -0.047 |
okay | -0.045 |
sorry | -0.045 |
lucky | -0.044 |
Token | him |
Token ID | 683 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
bons | +0.024 |
talk | +0.021 |
growth | +0.021 |
bars | +0.020 |
ests | +0.019 |
Neg logit contributions | |
glad | -0.030 |
pleased | -0.028 |
okay | -0.026 |
impressed | -0.026 |
lucky | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
glad | +0.058 |
pleased | +0.056 |
impressed | +0.052 |
lucky | +0.049 |
sorry | +0.049 |
Neg logit contributions | |
bons | -0.039 |
bars | -0.036 |
talk | -0.035 |
growth | -0.034 |
ales | -0.031 |
Token | The |
Token ID | 383 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
glad | +0.004 |
pleased | +0.004 |
impressed | +0.004 |
lucky | +0.004 |
okay | +0.004 |
Neg logit contributions | |
bons | -0.003 |
talk | -0.003 |
growth | -0.003 |
bars | -0.003 |
ests | -0.003 |
Token | end |
Token ID | 886 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
glad | +0.020 |
pleased | +0.020 |
impressed | +0.018 |
okay | +0.018 |
sorry | +0.018 |
Neg logit contributions | |
bons | -0.011 |
bars | -0.009 |
growth | -0.009 |
talk | -0.009 |
pages | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
glad | +0.024 |
pleased | +0.023 |
impressed | +0.021 |
lucky | +0.020 |
okay | +0.020 |
Neg logit contributions | |
bons | -0.017 |
talk | -0.015 |
bars | -0.015 |
growth | -0.015 |
ests | -0.013 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
glad | +0.014 |
pleased | +0.013 |
impressed | +0.012 |
sorry | +0.012 |
lucky | +0.012 |
Neg logit contributions | |
bons | -0.016 |
growth | -0.014 |
talk | -0.014 |
bars | -0.014 |
ests | -0.013 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
glad | +0.006 |
pleased | +0.005 |
impressed | +0.005 |
okay | +0.005 |
sorry | +0.005 |
Neg logit contributions | |
bons | -0.007 |
talk | -0.006 |
growth | -0.006 |
bars | -0.006 |
ests | -0.006 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
pleased | +0.146 |
glad | +0.141 |
sorry | +0.131 |
impressed | +0.131 |
okay | +0.122 |
Neg logit contributions | |
bons | -0.171 |
boxes | -0.170 |
talk | -0.168 |
ales | -0.166 |
board | -0.166 |
Token | a |
Token ID | 257 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
glad | +0.035 |
pleased | +0.034 |
impressed | +0.031 |
sorry | +0.030 |
lucky | +0.030 |
Neg logit contributions | |
bons | -0.031 |
growth | -0.028 |
talk | -0.027 |
bars | -0.027 |
shirts | -0.025 |
Token | loo |
Token ID | 29680 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
bons | +0.038 |
growth | +0.034 |
talk | +0.033 |
bars | +0.031 |
shirts | +0.031 |
Neg logit contributions | |
glad | -0.040 |
pleased | -0.037 |
lucky | -0.034 |
okay | -0.034 |
sorry | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
bons | +0.066 |
growth | +0.059 |
talk | +0.059 |
ests | +0.055 |
shirts | +0.054 |
Neg logit contributions | |
glad | -0.057 |
pleased | -0.053 |
okay | -0.049 |
lucky | -0.049 |
impressed | -0.048 |
Token | He |
Token ID | 679 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
bons | +0.074 |
talk | +0.069 |
growth | +0.064 |
ests | +0.061 |
els | +0.061 |
Neg logit contributions | |
glad | -0.075 |
pleased | -0.069 |
sorry | -0.065 |
okay | -0.064 |
lucky | -0.064 |
Token | thought |
Token ID | 1807 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
bons | +0.015 |
talk | +0.013 |
growth | +0.013 |
bars | +0.012 |
ests | +0.012 |
Neg logit contributions | |
glad | -0.016 |
pleased | -0.015 |
impressed | -0.013 |
okay | -0.013 |
sorry | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
glad | +0.039 |
pleased | +0.037 |
impressed | +0.034 |
okay | +0.032 |
sorry | +0.032 |
Neg logit contributions | |
bons | -0.038 |
growth | -0.035 |
bars | -0.035 |
shirts | -0.033 |
talk | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
bons | +0.016 |
talk | +0.014 |
growth | +0.014 |
bars | +0.014 |
ests | +0.013 |
Neg logit contributions | |
glad | -0.017 |
pleased | -0.016 |
impressed | -0.015 |
okay | -0.015 |
lucky | -0.015 |
Token | very |
Token ID | 845 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
glad | +0.053 |
pleased | +0.051 |
impressed | +0.045 |
sorry | +0.044 |
okay | +0.041 |
Neg logit contributions | |
bons | -0.065 |
cans | -0.056 |
bars | -0.056 |
boxes | -0.055 |
growth | -0.054 |
Token | cool |
Token ID | 3608 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
glad | +0.081 |
pleased | +0.077 |
impressed | +0.071 |
sorry | +0.071 |
okay | +0.070 |
Neg logit contributions | |
bons | -0.072 |
cans | -0.063 |
boxes | -0.062 |
ales | -0.062 |
talk | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
bons | +0.022 |
talk | +0.019 |
growth | +0.019 |
bars | +0.018 |
ests | +0.017 |
Neg logit contributions | |
glad | -0.024 |
pleased | -0.022 |
impressed | -0.021 |
okay | -0.021 |
lucky | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
glad | +0.068 |
pleased | +0.064 |
okay | +0.061 |
sorry | +0.060 |
impressed | +0.057 |
Neg logit contributions | |
bons | -0.058 |
bars | -0.050 |
cans | -0.049 |
shirts | -0.048 |
talk | -0.047 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
glad | +0.001 |
pleased | +0.001 |
sorry | +0.001 |
okay | +0.001 |
impressed | +0.001 |
Neg logit contributions | |
bons | -0.001 |
talk | -0.001 |
growth | -0.001 |
bars | -0.001 |
shirts | -0.001 |
Token | busy |
Token ID | 8179 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
glad | +0.111 |
pleased | +0.109 |
okay | +0.096 |
impressed | +0.091 |
lucky | +0.091 |
Neg logit contributions | |
bons | -0.123 |
shirts | -0.121 |
bars | -0.114 |
talk | -0.111 |
shelves | -0.106 |
Token | having |
Token ID | 1719 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
glad | +0.027 |
pleased | +0.026 |
okay | +0.023 |
impressed | +0.023 |
lucky | +0.022 |
Neg logit contributions | |
bons | -0.040 |
talk | -0.036 |
growth | -0.036 |
bars | -0.035 |
shirts | -0.034 |
Token | fun |
Token ID | 1257 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
bons | +0.025 |
talk | +0.022 |
growth | +0.022 |
bars | +0.021 |
ests | +0.020 |
Neg logit contributions | |
glad | -0.026 |
pleased | -0.025 |
impressed | -0.023 |
okay | -0.023 |
lucky | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +5.71e-03 |
Pos logit contributions | |
glad | +0.018 |
pleased | +0.017 |
okay | +0.015 |
impressed | +0.015 |
lucky | +0.015 |
Neg logit contributions | |
bons | -0.025 |
growth | -0.023 |
talk | -0.022 |
bars | -0.022 |
shirts | -0.022 |
Token | something |
Token ID | 1223 |
Feature activation | +7.85e-01 |
Pos logit contributions | |
glad | +0.000 |
pleased | +0.000 |
impressed | +0.000 |
okay | +0.000 |
sorry | +0.000 |
Neg logit contributions | |
bons | -0.000 |
growth | -0.000 |
talk | -0.000 |
bars | -0.000 |
shirts | -0.000 |
Token | very |
Token ID | 845 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
glad | +0.035 |
pleased | +0.031 |
okay | +0.028 |
impressed | +0.027 |
sorry | +0.027 |
Neg logit contributions | |
bons | -0.050 |
growth | -0.046 |
bars | -0.046 |
talk | -0.046 |
shirts | -0.044 |
Token | special |
Token ID | 2041 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
glad | +0.115 |
okay | +0.097 |
pleased | +0.095 |
shadow | +0.089 |
impressed | +0.085 |
Neg logit contributions | |
talk | -0.137 |
papers | -0.132 |
bons | -0.131 |
bars | -0.130 |
shirts | -0.130 |
Token | happened |
Token ID | 3022 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
bons | +0.087 |
talk | +0.079 |
growth | +0.079 |
bars | +0.076 |
ales | +0.074 |
Neg logit contributions | |
glad | -0.036 |
pleased | -0.032 |
lucky | -0.028 |
okay | -0.027 |
impressed | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
glad | +0.106 |
pleased | +0.102 |
impressed | +0.094 |
sorry | +0.093 |
lucky | +0.093 |
Neg logit contributions | |
bons | -0.054 |
talk | -0.044 |
bars | -0.043 |
growth | -0.042 |
shirts | -0.042 |
Token | Tracy |
Token ID | 31140 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
bons | +0.045 |
talk | +0.040 |
growth | +0.038 |
bars | +0.037 |
ests | +0.036 |
Neg logit contributions | |
glad | -0.053 |
pleased | -0.050 |
lucky | -0.046 |
sorry | -0.046 |
impressed | -0.046 |
Token | went |
Token ID | 1816 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
bons | +0.022 |
talk | +0.020 |
growth | +0.020 |
bars | +0.019 |
ests | +0.018 |
Neg logit contributions | |
glad | -0.024 |
pleased | -0.022 |
impressed | -0.020 |
okay | -0.020 |
lucky | -0.020 |
Token | They |
Token ID | 1119 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
bons | +0.051 |
talk | +0.043 |
growth | +0.041 |
ests | +0.039 |
bars | +0.039 |
Neg logit contributions | |
glad | -0.079 |
pleased | -0.073 |
sorry | -0.070 |
okay | -0.069 |
lucky | -0.068 |
Token | wonder |
Token ID | 4240 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
bons | +0.103 |
ests | +0.092 |
growth | +0.089 |
talk | +0.089 |
orns | +0.086 |
Neg logit contributions | |
glad | -0.091 |
pleased | -0.083 |
sorry | -0.079 |
lucky | -0.078 |
okay | -0.078 |
Token | what |
Token ID | 644 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
bons | +0.029 |
talk | +0.026 |
growth | +0.026 |
bars | +0.025 |
ests | +0.025 |
Neg logit contributions | |
glad | -0.016 |
pleased | -0.015 |
impressed | -0.013 |
okay | -0.013 |
lucky | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
bons | +0.034 |
talk | +0.030 |
growth | +0.030 |
bars | +0.029 |
ales | +0.028 |
Neg logit contributions | |
glad | -0.033 |
pleased | -0.031 |
lucky | -0.028 |
impressed | -0.028 |
okay | -0.028 |
Token | surprise |
Token ID | 5975 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
bons | +0.030 |
talk | +0.026 |
growth | +0.026 |
bars | +0.025 |
ales | +0.024 |
Neg logit contributions | |
glad | -0.034 |
pleased | -0.032 |
lucky | -0.029 |
okay | -0.029 |
impressed | -0.029 |
Token | is |
Token ID | 318 |
Feature activation | +6.78e-01 |
Pos logit contributions | |
bons | +0.116 |
growth | +0.106 |
talk | +0.104 |
ales | +0.096 |
ests | +0.094 |
Neg logit contributions | |
glad | -0.092 |
pleased | -0.084 |
lucky | -0.080 |
okay | -0.079 |
sorry | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
glad | +0.044 |
pleased | +0.042 |
impressed | +0.038 |
sorry | +0.038 |
okay | +0.037 |
Neg logit contributions | |
bons | -0.031 |
talk | -0.027 |
bars | -0.026 |
growth | -0.025 |
cans | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
bons | +0.037 |
talk | +0.031 |
growth | +0.029 |
bars | +0.029 |
ales | +0.028 |
Neg logit contributions | |
glad | -0.062 |
pleased | -0.058 |
okay | -0.055 |
sorry | -0.055 |
lucky | -0.054 |
Token | hope |
Token ID | 2911 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
bons | +0.125 |
ests | +0.112 |
talk | +0.106 |
growth | +0.106 |
orns | +0.105 |
Neg logit contributions | |
glad | -0.106 |
pleased | -0.096 |
lucky | -0.092 |
sorry | -0.090 |
okay | -0.089 |
Token | it |
Token ID | 340 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
bons | +0.011 |
talk | +0.010 |
growth | +0.010 |
bars | +0.009 |
shirts | +0.009 |
Neg logit contributions | |
glad | -0.010 |
pleased | -0.009 |
impressed | -0.009 |
okay | -0.008 |
lucky | -0.008 |
Token | is |
Token ID | 318 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
bons | +0.063 |
growth | +0.056 |
talk | +0.055 |
ales | +0.052 |
ests | +0.052 |
Neg logit contributions | |
glad | -0.058 |
pleased | -0.053 |
lucky | -0.050 |
okay | -0.049 |
sorry | -0.049 |
Token | sleep |
Token ID | 3993 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
glad | +0.028 |
pleased | +0.026 |
impressed | +0.024 |
lucky | +0.024 |
okay | +0.023 |
Neg logit contributions | |
bons | -0.022 |
talk | -0.019 |
growth | -0.019 |
bars | -0.018 |
shirts | -0.017 |
Token | ." |
Token ID | 526 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
glad | +0.110 |
pleased | +0.106 |
impressed | +0.097 |
lucky | +0.091 |
sorry | +0.086 |
Neg logit contributions | |
bons | -0.105 |
growth | -0.088 |
bars | -0.088 |
talk | -0.085 |
shirts | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
glad | +0.013 |
pleased | +0.012 |
impressed | +0.011 |
lucky | +0.011 |
okay | +0.011 |
Neg logit contributions | |
bons | -0.014 |
talk | -0.012 |
growth | -0.012 |
bars | -0.012 |
shirts | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
glad | +0.028 |
pleased | +0.026 |
lucky | +0.024 |
okay | +0.024 |
sorry | +0.024 |
Neg logit contributions | |
bons | -0.031 |
talk | -0.028 |
growth | -0.027 |
bars | -0.026 |
ests | -0.025 |
Token | The |
Token ID | 464 |
Feature activation | +6.36e-01 |
Pos logit contributions | |
bons | +0.020 |
talk | +0.018 |
growth | +0.017 |
bars | +0.017 |
ests | +0.016 |
Neg logit contributions | |
glad | -0.024 |
pleased | -0.023 |
impressed | -0.021 |
okay | -0.021 |
lucky | -0.021 |
Token | lion |
Token ID | 18744 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
glad | +0.037 |
pleased | +0.035 |
impressed | +0.032 |
okay | +0.031 |
lucky | +0.030 |
Neg logit contributions | |
bons | -0.033 |
talk | -0.029 |
growth | -0.029 |
bars | -0.029 |
shirts | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
bons | +0.002 |
talk | +0.001 |
growth | +0.001 |
bars | +0.001 |
shirts | +0.001 |
Neg logit contributions | |
glad | -0.002 |
pleased | -0.002 |
okay | -0.001 |
impressed | -0.001 |
lucky | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
glad | +0.148 |
pleased | +0.131 |
impressed | +0.125 |
lucky | +0.125 |
shadow | +0.112 |
Neg logit contributions | |
als | -0.105 |
bons | -0.104 |
bars | -0.102 |
talk | -0.102 |
board | -0.100 |
Token | " |
Token ID | 366 |
Feature activation | -9.56e-01 |
Pos logit contributions | |
glad | +0.114 |
pleased | +0.106 |
lucky | +0.101 |
impressed | +0.096 |
okay | +0.095 |
Neg logit contributions | |
bons | -0.115 |
growth | -0.106 |
bars | -0.105 |
als | -0.103 |
talk | -0.103 |
Token | I |
Token ID | 40 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
bons | +0.050 |
growth | +0.043 |
talk | +0.043 |
bars | +0.043 |
ales | +0.041 |
Neg logit contributions | |
glad | -0.057 |
pleased | -0.054 |
impressed | -0.049 |
okay | -0.049 |
lucky | -0.049 |
Token | can |
Token ID | 460 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
glad | +0.075 |
pleased | +0.071 |
impressed | +0.068 |
sorry | +0.064 |
lucky | +0.063 |
Neg logit contributions | |
bons | -0.054 |
talk | -0.054 |
bars | -0.049 |
growth | -0.046 |
ests | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
bons | +0.022 |
talk | +0.020 |
growth | +0.020 |
bars | +0.019 |
ests | +0.019 |
Neg logit contributions | |
glad | -0.018 |
pleased | -0.016 |
impressed | -0.015 |
okay | -0.015 |
lucky | -0.014 |
Token | burn |
Token ID | 4245 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
bons | +0.051 |
talk | +0.046 |
growth | +0.044 |
ests | +0.044 |
bars | +0.043 |
Neg logit contributions | |
glad | -0.058 |
pleased | -0.055 |
okay | -0.051 |
impressed | -0.051 |
sorry | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -8.01e-01 |
Pos logit contributions | |
bons | +0.091 |
talk | +0.086 |
growth | +0.083 |
ests | +0.078 |
bars | +0.076 |
Neg logit contributions | |
glad | -0.073 |
pleased | -0.068 |
okay | -0.065 |
lucky | -0.063 |
sorry | -0.063 |
Token | |
Token ID | 220 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
bons | +0.046 |
talk | +0.042 |
growth | +0.041 |
bars | +0.039 |
ales | +0.039 |
Neg logit contributions | |
glad | -0.044 |
pleased | -0.041 |
okay | -0.037 |
impressed | -0.037 |
lucky | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
glad | +0.055 |
pleased | +0.053 |
impressed | +0.048 |
sorry | +0.048 |
lucky | +0.047 |
Neg logit contributions | |
bons | -0.054 |
talk | -0.048 |
growth | -0.047 |
bars | -0.047 |
ests | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
glad | +0.011 |
pleased | +0.010 |
sorry | +0.010 |
lucky | +0.009 |
impressed | +0.009 |
Neg logit contributions | |
bons | -0.013 |
talk | -0.012 |
growth | -0.011 |
bars | -0.011 |
ests | -0.011 |
Token | Tom |
Token ID | 13787 |
Feature activation | -9.07e-01 |
Pos logit contributions | |
bons | +0.017 |
talk | +0.014 |
growth | +0.014 |
bars | +0.013 |
ests | +0.013 |
Neg logit contributions | |
glad | -0.024 |
pleased | -0.022 |
impressed | -0.021 |
okay | -0.021 |
lucky | -0.021 |
Token | my |
Token ID | 1820 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
bons | +0.046 |
growth | +0.041 |
talk | +0.040 |
bars | +0.037 |
ests | +0.037 |
Neg logit contributions | |
glad | -0.056 |
pleased | -0.053 |
impressed | -0.049 |
okay | -0.048 |
lucky | -0.048 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
bons | +0.009 |
talk | +0.008 |
growth | +0.008 |
bars | +0.008 |
shirts | +0.007 |
Neg logit contributions | |
glad | -0.010 |
pleased | -0.009 |
okay | -0.008 |
impressed | -0.008 |
lucky | -0.008 |
Token | very |
Token ID | 845 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
glad | +0.045 |
pleased | +0.043 |
impressed | +0.038 |
okay | +0.036 |
shadow | +0.036 |
Neg logit contributions | |
bons | -0.081 |
growth | -0.077 |
bars | -0.076 |
cans | -0.074 |
talk | -0.071 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
glad | +0.061 |
shadow | +0.061 |
reetings | +0.060 |
pleased | +0.053 |
okay | +0.050 |
Neg logit contributions | |
bars | -0.135 |
cans | -0.135 |
tips | -0.134 |
leaves | -0.133 |
bons | -0.130 |
Token | started |
Token ID | 2067 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
glad | +0.014 |
pleased | +0.013 |
okay | +0.012 |
sorry | +0.012 |
impressed | +0.012 |
Neg logit contributions | |
bons | -0.013 |
talk | -0.012 |
growth | -0.012 |
bars | -0.011 |
shirts | -0.011 |
Token | walking |
Token ID | 6155 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
glad | +0.026 |
pleased | +0.024 |
lucky | +0.022 |
okay | +0.022 |
impressed | +0.022 |
Neg logit contributions | |
bons | -0.029 |
talk | -0.026 |
growth | -0.026 |
bars | -0.026 |
shirts | -0.024 |
Token | with |
Token ID | 351 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
bons | +0.018 |
talk | +0.016 |
growth | +0.016 |
bars | +0.015 |
ests | +0.015 |
Neg logit contributions | |
glad | -0.020 |
pleased | -0.019 |
impressed | -0.017 |
okay | -0.017 |
lucky | -0.017 |
Token | his |
Token ID | 465 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
bons | +0.055 |
talk | +0.051 |
growth | +0.048 |
bars | +0.047 |
ests | +0.046 |
Neg logit contributions | |
glad | -0.055 |
pleased | -0.051 |
lucky | -0.047 |
okay | -0.047 |
sorry | -0.047 |
Token | head |
Token ID | 1182 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
bons | +0.007 |
talk | +0.006 |
growth | +0.006 |
bars | +0.006 |
shirts | +0.006 |
Neg logit contributions | |
glad | -0.007 |
pleased | -0.006 |
impressed | -0.006 |
okay | -0.006 |
sorry | -0.005 |
Token | down |
Token ID | 866 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
glad | +0.065 |
pleased | +0.061 |
impressed | +0.056 |
lucky | +0.055 |
sorry | +0.055 |
Neg logit contributions | |
bons | -0.058 |
talk | -0.052 |
bars | -0.051 |
growth | -0.050 |
shirts | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
bons | +0.050 |
talk | +0.045 |
growth | +0.044 |
bars | +0.042 |
ests | +0.042 |
Neg logit contributions | |
glad | -0.042 |
pleased | -0.039 |
impressed | -0.035 |
okay | -0.035 |
lucky | -0.035 |
Token | He |
Token ID | 679 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
bons | +0.025 |
talk | +0.022 |
growth | +0.021 |
bars | +0.020 |
ests | +0.020 |
Neg logit contributions | |
glad | -0.031 |
pleased | -0.030 |
impressed | -0.027 |
okay | -0.027 |
lucky | -0.027 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
glad | +0.022 |
pleased | +0.021 |
sorry | +0.019 |
okay | +0.019 |
impressed | +0.019 |
Neg logit contributions | |
bons | -0.019 |
talk | -0.018 |
growth | -0.017 |
bars | -0.017 |
shirts | -0.016 |
Token | up |
Token ID | 510 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
glad | +0.146 |
pleased | +0.137 |
okay | +0.131 |
sorry | +0.127 |
impressed | +0.127 |
Neg logit contributions | |
bons | -0.089 |
bars | -0.082 |
ted | -0.082 |
boxes | -0.081 |
cans | -0.080 |
Token | and |
Token ID | 290 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
glad | +0.098 |
pleased | +0.091 |
sorry | +0.088 |
okay | +0.086 |
lucky | +0.085 |
Neg logit contributions | |
bons | -0.063 |
growth | -0.058 |
talk | -0.057 |
shirts | -0.053 |
als | -0.053 |
Token | " |
Token ID | 366 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
pleased | +0.077 |
glad | +0.076 |
lucky | +0.068 |
impressed | +0.068 |
okay | +0.062 |
Neg logit contributions | |
bons | -0.086 |
bars | -0.080 |
talk | -0.078 |
cans | -0.078 |
growth | -0.077 |
Token | I |
Token ID | 40 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
bons | +0.075 |
growth | +0.069 |
talk | +0.066 |
bars | +0.066 |
ales | +0.064 |
Neg logit contributions | |
glad | -0.067 |
pleased | -0.061 |
impressed | -0.057 |
okay | -0.057 |
lucky | -0.056 |
Token | 'm |
Token ID | 1101 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
glad | +0.005 |
pleased | +0.005 |
impressed | +0.004 |
sorry | +0.004 |
lucky | +0.004 |
Neg logit contributions | |
bons | -0.007 |
talk | -0.006 |
growth | -0.006 |
bars | -0.006 |
ests | -0.006 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
pleased | +0.050 |
reetings | +0.044 |
mber | +0.034 |
impressed | +0.033 |
upon | +0.031 |
Neg logit contributions | |
bons | -0.149 |
bars | -0.145 |
cans | -0.140 |
boxes | -0.136 |
newspapers | -0.133 |
Token | your |
Token ID | 534 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
pleased | +0.083 |
glad | +0.081 |
impressed | +0.074 |
lucky | +0.066 |
thankful | +0.065 |
Neg logit contributions | |
bons | -0.065 |
shirts | -0.059 |
cans | -0.059 |
bars | -0.059 |
talk | -0.056 |
Token | cars |
Token ID | 5006 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
glad | +0.025 |
pleased | +0.024 |
impressed | +0.023 |
okay | +0.022 |
lucky | +0.022 |
Neg logit contributions | |
bons | -0.013 |
talk | -0.011 |
bars | -0.011 |
shirts | -0.011 |
ests | -0.010 |
Token | are |
Token ID | 389 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
bons | +0.025 |
talk | +0.022 |
growth | +0.022 |
bars | +0.021 |
ests | +0.020 |
Neg logit contributions | |
glad | -0.027 |
pleased | -0.025 |
okay | -0.023 |
impressed | -0.023 |
lucky | -0.023 |
Token | broken |
Token ID | 5445 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
glad | +0.077 |
pleased | +0.077 |
impressed | +0.064 |
mber | +0.062 |
reetings | +0.058 |
Neg logit contributions | |
bons | -0.122 |
bars | -0.107 |
tape | -0.105 |
cans | -0.104 |
shirts | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
bons | +0.006 |
talk | +0.005 |
growth | +0.005 |
bars | +0.005 |
shirts | +0.005 |
Neg logit contributions | |
glad | -0.007 |
pleased | -0.006 |
impressed | -0.006 |
lucky | -0.006 |
okay | -0.006 |
Token | But |
Token ID | 887 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
bons | +0.044 |
talk | +0.040 |
growth | +0.039 |
bars | +0.038 |
ales | +0.037 |
Neg logit contributions | |
glad | -0.034 |
pleased | -0.031 |
sorry | -0.029 |
okay | -0.029 |
lucky | -0.028 |
Token | you |
Token ID | 345 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
pleased | +0.061 |
glad | +0.060 |
impressed | +0.054 |
lucky | +0.049 |
okay | +0.047 |
Neg logit contributions | |
bons | -0.067 |
cans | -0.059 |
shirts | -0.057 |
bars | -0.057 |
talk | -0.056 |
Token | He |
Token ID | 679 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
bons | +0.085 |
talk | +0.080 |
growth | +0.075 |
bars | +0.072 |
els | +0.070 |
Neg logit contributions | |
glad | -0.084 |
pleased | -0.075 |
okay | -0.074 |
lucky | -0.071 |
sorry | -0.070 |
Token | would |
Token ID | 561 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
bons | +0.067 |
growth | +0.061 |
talk | +0.061 |
ales | +0.057 |
ests | +0.056 |
Neg logit contributions | |
glad | -0.052 |
pleased | -0.047 |
lucky | -0.044 |
impressed | -0.044 |
okay | -0.042 |
Token | paint |
Token ID | 7521 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
bons | +0.070 |
talk | +0.064 |
growth | +0.063 |
ests | +0.062 |
ales | +0.057 |
Neg logit contributions | |
glad | -0.093 |
pleased | -0.086 |
sorry | -0.081 |
impressed | -0.081 |
lucky | -0.081 |
Token | all |
Token ID | 477 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
bons | +0.108 |
talk | +0.108 |
growth | +0.106 |
ests | +0.095 |
ales | +0.095 |
Neg logit contributions | |
glad | -0.095 |
pleased | -0.087 |
lucky | -0.085 |
okay | -0.083 |
impressed | -0.082 |
Token | day |
Token ID | 1110 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
bons | +0.017 |
talk | +0.015 |
growth | +0.015 |
bars | +0.014 |
ests | +0.014 |
Neg logit contributions | |
glad | -0.016 |
pleased | -0.015 |
impressed | -0.014 |
okay | -0.014 |
lucky | -0.014 |
Token | long |
Token ID | 890 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
bons | +0.001 |
talk | +0.001 |
growth | +0.001 |
bars | +0.001 |
shirts | +0.001 |
Neg logit contributions | |
glad | -0.001 |
pleased | -0.001 |
okay | -0.000 |
sorry | -0.000 |
impressed | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
bons | +0.098 |
talk | +0.093 |
growth | +0.089 |
ests | +0.083 |
ales | +0.082 |
Neg logit contributions | |
glad | -0.074 |
pleased | -0.066 |
okay | -0.062 |
impressed | -0.062 |
lucky | -0.061 |
Token | His |
Token ID | 2399 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
bons | +0.048 |
talk | +0.044 |
growth | +0.042 |
bars | +0.041 |
shirts | +0.040 |
Neg logit contributions | |
glad | -0.039 |
pleased | -0.035 |
okay | -0.033 |
lucky | -0.032 |
sorry | -0.032 |
Token | favorite |
Token ID | 4004 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
bons | +0.029 |
talk | +0.026 |
growth | +0.026 |
bars | +0.025 |
ales | +0.024 |
Neg logit contributions | |
glad | -0.022 |
pleased | -0.020 |
lucky | -0.018 |
okay | -0.018 |
impressed | -0.018 |
Token | thing |
Token ID | 1517 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
bons | +0.090 |
talk | +0.089 |
growth | +0.086 |
ales | +0.081 |
ests | +0.073 |
Neg logit contributions | |
glad | -0.097 |
pleased | -0.089 |
lucky | -0.088 |
happy | -0.084 |
okay | -0.081 |
Token | to |
Token ID | 284 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
bons | +0.063 |
talk | +0.058 |
growth | +0.057 |
ales | +0.053 |
bars | +0.053 |
Neg logit contributions | |
glad | -0.046 |
pleased | -0.042 |
lucky | -0.038 |
impressed | -0.038 |
okay | -0.038 |
Token | revealed |
Token ID | 4602 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
bons | +0.057 |
talk | +0.050 |
growth | +0.050 |
ests | +0.047 |
bars | +0.047 |
Neg logit contributions | |
glad | -0.052 |
pleased | -0.049 |
impressed | -0.045 |
lucky | -0.044 |
okay | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
bons | +0.052 |
talk | +0.047 |
growth | +0.046 |
bars | +0.044 |
ests | +0.044 |
Neg logit contributions | |
glad | -0.044 |
pleased | -0.041 |
okay | -0.038 |
lucky | -0.037 |
sorry | -0.037 |
Token | big |
Token ID | 1263 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
glad | +0.099 |
pleased | +0.090 |
okay | +0.087 |
sorry | +0.083 |
impressed | +0.082 |
Neg logit contributions | |
bons | -0.093 |
bars | -0.087 |
papers | -0.084 |
board | -0.084 |
shirts | -0.082 |
Token | gray |
Token ID | 12768 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
bons | +0.006 |
talk | +0.005 |
growth | +0.005 |
bars | +0.005 |
shirts | +0.005 |
Neg logit contributions | |
glad | -0.009 |
pleased | -0.008 |
okay | -0.008 |
impressed | -0.008 |
lucky | -0.007 |
Token | bunny |
Token ID | 44915 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
bons | +0.076 |
talk | +0.068 |
growth | +0.064 |
ales | +0.061 |
ests | +0.058 |
Neg logit contributions | |
glad | -0.105 |
pleased | -0.099 |
lucky | -0.094 |
sorry | -0.093 |
impressed | -0.091 |
Token | on |
Token ID | 319 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
bons | +0.038 |
talk | +0.034 |
growth | +0.034 |
bars | +0.032 |
ests | +0.031 |
Neg logit contributions | |
glad | -0.035 |
pleased | -0.033 |
lucky | -0.030 |
impressed | -0.030 |
okay | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
talk | +0.160 |
bons | +0.154 |
ests | +0.151 |
ales | +0.144 |
bars | +0.137 |
Neg logit contributions | |
glad | -0.097 |
pleased | -0.086 |
lucky | -0.083 |
sorry | -0.082 |
impressed | -0.079 |
Token | black |
Token ID | 2042 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
bons | +0.093 |
talk | +0.087 |
ests | +0.080 |
growth | +0.080 |
orns | +0.076 |
Neg logit contributions | |
glad | -0.122 |
lucky | -0.114 |
pleased | -0.113 |
sorry | -0.106 |
happy | -0.104 |
Token | board |
Token ID | 3526 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
bons | +0.107 |
talk | +0.096 |
stations | +0.086 |
ales | +0.085 |
aled | +0.076 |
Neg logit contributions | |
glad | -0.201 |
pleased | -0.195 |
lucky | -0.195 |
sorry | -0.195 |
okay | -0.189 |
Token | ! |
Token ID | 0 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
bons | +0.119 |
growth | +0.109 |
talk | +0.107 |
ales | +0.102 |
ests | +0.099 |
Neg logit contributions | |
glad | -0.099 |
pleased | -0.093 |
impressed | -0.085 |
sorry | -0.085 |
okay | -0.084 |
Token | |
Token ID | 220 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
bons | +0.046 |
talk | +0.041 |
growth | +0.041 |
bars | +0.039 |
ales | +0.039 |
Neg logit contributions | |
glad | -0.039 |
pleased | -0.036 |
impressed | -0.032 |
sorry | -0.032 |
okay | -0.032 |
Token | us |
Token ID | 514 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
bons | +0.038 |
talk | +0.035 |
growth | +0.034 |
bars | +0.033 |
ests | +0.032 |
Neg logit contributions | |
glad | -0.029 |
pleased | -0.026 |
okay | -0.024 |
impressed | -0.024 |
lucky | -0.024 |
Token | ." |
Token ID | 526 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
glad | +0.022 |
pleased | +0.021 |
impressed | +0.020 |
sorry | +0.019 |
lucky | +0.019 |
Neg logit contributions | |
bons | -0.020 |
talk | -0.018 |
bars | -0.018 |
growth | -0.017 |
shirts | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
bons | +0.110 |
growth | +0.101 |
ales | +0.098 |
talk | +0.097 |
els | +0.094 |
Neg logit contributions | |
glad | -0.061 |
pleased | -0.054 |
sorry | -0.049 |
okay | -0.048 |
lucky | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
bons | +0.004 |
talk | +0.004 |
growth | +0.004 |
bars | +0.003 |
ests | +0.003 |
Neg logit contributions | |
glad | -0.003 |
pleased | -0.003 |
okay | -0.003 |
impressed | -0.003 |
lucky | -0.003 |
Token | L |
Token ID | 43 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
bons | +0.075 |
growth | +0.072 |
ales | +0.065 |
talk | +0.065 |
als | +0.064 |
Neg logit contributions | |
glad | -0.073 |
pleased | -0.070 |
impressed | -0.065 |
okay | -0.064 |
lucky | -0.063 |
Token | ily |
Token ID | 813 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
bons | +0.072 |
growth | +0.067 |
talk | +0.066 |
als | +0.060 |
bars | +0.059 |
Neg logit contributions | |
glad | -0.074 |
pleased | -0.068 |
okay | -0.066 |
sorry | -0.064 |
lucky | -0.063 |
Token | says |
Token ID | 1139 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
bons | +0.142 |
growth | +0.134 |
talk | +0.125 |
ales | +0.122 |
ests | +0.117 |
Neg logit contributions | |
glad | -0.132 |
pleased | -0.120 |
sorry | -0.115 |
impressed | -0.115 |
lucky | -0.113 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
glad | +0.056 |
pleased | +0.054 |
impressed | +0.051 |
lucky | +0.047 |
thankful | +0.043 |
Neg logit contributions | |
bons | -0.056 |
talk | -0.052 |
bars | -0.052 |
shirts | -0.050 |
als | -0.050 |
Token | " |
Token ID | 366 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
glad | +0.061 |
pleased | +0.059 |
impressed | +0.054 |
lucky | +0.053 |
okay | +0.051 |
Neg logit contributions | |
bons | -0.068 |
talk | -0.061 |
bars | -0.061 |
als | -0.060 |
growth | -0.060 |
Token | Thank |
Token ID | 10449 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
bons | +0.081 |
growth | +0.076 |
els | +0.071 |
ales | +0.071 |
bars | +0.070 |
Neg logit contributions | |
glad | -0.084 |
pleased | -0.076 |
lucky | -0.070 |
okay | -0.069 |
impressed | -0.068 |
Token | you |
Token ID | 345 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
bons | +0.073 |
growth | +0.068 |
talk | +0.068 |
bars | +0.065 |
ales | +0.063 |
Neg logit contributions | |
glad | -0.039 |
pleased | -0.034 |
okay | -0.032 |
sorry | -0.032 |
lucky | -0.031 |
Token | trunk |
Token ID | 21427 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
bons | +0.055 |
growth | +0.050 |
talk | +0.049 |
ales | +0.047 |
als | +0.042 |
Neg logit contributions | |
glad | -0.088 |
pleased | -0.080 |
sorry | -0.079 |
lucky | -0.079 |
okay | -0.078 |
Token | is |
Token ID | 318 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
bons | +0.131 |
growth | +0.122 |
talk | +0.115 |
ales | +0.111 |
ests | +0.105 |
Neg logit contributions | |
glad | -0.111 |
pleased | -0.102 |
sorry | -0.098 |
lucky | -0.096 |
okay | -0.094 |
Token | a |
Token ID | 257 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
glad | +0.067 |
pleased | +0.064 |
impressed | +0.056 |
lucky | +0.054 |
okay | +0.051 |
Neg logit contributions | |
bons | -0.094 |
boxes | -0.085 |
cans | -0.084 |
papers | -0.083 |
shirts | -0.083 |
Token | big |
Token ID | 1263 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
glad | +0.077 |
pleased | +0.073 |
impressed | +0.068 |
okay | +0.066 |
sorry | +0.062 |
Neg logit contributions | |
bons | -0.075 |
bars | -0.068 |
cans | -0.066 |
boxes | -0.065 |
growth | -0.065 |
Token | box |
Token ID | 3091 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
bons | +0.016 |
talk | +0.014 |
growth | +0.014 |
bars | +0.013 |
ests | +0.013 |
Neg logit contributions | |
glad | -0.022 |
pleased | -0.020 |
impressed | -0.019 |
okay | -0.019 |
lucky | -0.019 |
Token | with |
Token ID | 351 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
glad | +0.006 |
pleased | +0.006 |
impressed | +0.005 |
okay | +0.005 |
lucky | +0.005 |
Neg logit contributions | |
bons | -0.006 |
talk | -0.005 |
growth | -0.005 |
bars | -0.005 |
shirts | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
bons | +0.136 |
talk | +0.134 |
growth | +0.126 |
ales | +0.122 |
ests | +0.121 |
Neg logit contributions | |
glad | -0.118 |
lucky | -0.105 |
pleased | -0.105 |
sorry | -0.100 |
okay | -0.098 |
Token | lid |
Token ID | 19789 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
glad | +0.082 |
pleased | +0.077 |
impressed | +0.072 |
okay | +0.071 |
lucky | +0.068 |
Neg logit contributions | |
bons | -0.060 |
bars | -0.053 |
growth | -0.050 |
talk | -0.049 |
shirts | -0.049 |
Token | that |
Token ID | 326 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
bons | +0.005 |
talk | +0.005 |
growth | +0.005 |
bars | +0.005 |
shirts | +0.004 |
Neg logit contributions | |
glad | -0.006 |
pleased | -0.006 |
impressed | -0.005 |
okay | -0.005 |
lucky | -0.005 |
Token | can |
Token ID | 460 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
bons | +0.061 |
growth | +0.053 |
talk | +0.053 |
ests | +0.049 |
ales | +0.049 |
Neg logit contributions | |
glad | -0.068 |
pleased | -0.062 |
sorry | -0.060 |
lucky | -0.059 |
okay | -0.059 |
Token | hold |
Token ID | 1745 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
bons | +0.070 |
talk | +0.061 |
growth | +0.060 |
ests | +0.057 |
bars | +0.055 |
Neg logit contributions | |
glad | -0.075 |
pleased | -0.070 |
okay | -0.067 |
sorry | -0.067 |
lucky | -0.066 |
Token | start |
Token ID | 923 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
bons | +0.083 |
growth | +0.075 |
talk | +0.071 |
ests | +0.071 |
ales | +0.067 |
Neg logit contributions | |
glad | -0.067 |
pleased | -0.060 |
sorry | -0.059 |
okay | -0.058 |
lucky | -0.057 |
Token | to |
Token ID | 284 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
bons | +0.027 |
talk | +0.025 |
growth | +0.025 |
bars | +0.024 |
ests | +0.023 |
Neg logit contributions | |
glad | -0.021 |
pleased | -0.019 |
impressed | -0.017 |
okay | -0.017 |
lucky | -0.017 |
Token | fight |
Token ID | 1907 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
bons | +0.027 |
talk | +0.023 |
growth | +0.023 |
bars | +0.022 |
ests | +0.021 |
Neg logit contributions | |
glad | -0.033 |
pleased | -0.031 |
okay | -0.029 |
impressed | -0.029 |
sorry | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
bons | +0.071 |
growth | +0.065 |
talk | +0.063 |
ests | +0.059 |
ales | +0.059 |
Neg logit contributions | |
glad | -0.061 |
pleased | -0.056 |
okay | -0.054 |
sorry | -0.053 |
impressed | -0.052 |
Token | They |
Token ID | 1119 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
bons | +0.062 |
talk | +0.054 |
growth | +0.052 |
ales | +0.051 |
bars | +0.051 |
Neg logit contributions | |
glad | -0.081 |
pleased | -0.073 |
okay | -0.072 |
sorry | -0.072 |
lucky | -0.069 |
Token | throw |
Token ID | 3714 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
bons | +0.092 |
growth | +0.082 |
ests | +0.080 |
talk | +0.077 |
orns | +0.075 |
Neg logit contributions | |
glad | -0.085 |
pleased | -0.077 |
sorry | -0.077 |
okay | -0.075 |
lucky | -0.073 |
Token | their |
Token ID | 511 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
bons | +0.202 |
growth | +0.198 |
talk | +0.197 |
ales | +0.186 |
ests | +0.186 |
Neg logit contributions | |
glad | -0.117 |
sad | -0.113 |
sorry | -0.112 |
happy | -0.109 |
okay | -0.108 |
Token | toys |
Token ID | 14958 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
bons | +0.085 |
growth | +0.079 |
talk | +0.079 |
ales | +0.072 |
ests | +0.066 |
Neg logit contributions | |
glad | -0.100 |
lucky | -0.091 |
pleased | -0.090 |
sorry | -0.090 |
okay | -0.087 |
Token | at |
Token ID | 379 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
bons | +0.163 |
growth | +0.152 |
talk | +0.143 |
ests | +0.138 |
ales | +0.133 |
Neg logit contributions | |
glad | -0.111 |
pleased | -0.100 |
okay | -0.100 |
sorry | -0.097 |
lucky | -0.094 |
Token | each |
Token ID | 1123 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
bons | +0.154 |
talk | +0.151 |
growth | +0.136 |
ests | +0.135 |
als | +0.134 |
Neg logit contributions | |
glad | -0.139 |
happy | -0.121 |
okay | -0.119 |
lucky | -0.118 |
sorry | -0.116 |
Token | other |
Token ID | 584 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
bons | +0.194 |
growth | +0.181 |
talk | +0.177 |
als | +0.174 |
ests | +0.168 |
Neg logit contributions | |
glad | -0.117 |
happy | -0.110 |
lucky | -0.104 |
sad | -0.100 |
pleased | -0.099 |
Token |
|
Token ID | 198 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
bons | +0.006 |
talk | +0.005 |
growth | +0.005 |
bars | +0.005 |
ests | +0.005 |
Neg logit contributions | |
glad | -0.007 |
pleased | -0.007 |
impressed | -0.006 |
lucky | -0.006 |
okay | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
glad | +0.024 |
pleased | +0.023 |
okay | +0.021 |
sorry | +0.021 |
lucky | +0.021 |
Neg logit contributions | |
bons | -0.026 |
talk | -0.023 |
growth | -0.023 |
bars | -0.022 |
ests | -0.021 |
Token | Tom |
Token ID | 13787 |
Feature activation | -9.33e-01 |
Pos logit contributions | |
bons | +0.005 |
talk | +0.005 |
growth | +0.005 |
bars | +0.004 |
shirts | +0.004 |
Neg logit contributions | |
glad | -0.008 |
pleased | -0.007 |
impressed | -0.007 |
okay | -0.007 |
lucky | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
bons | +0.048 |
growth | +0.041 |
talk | +0.041 |
ales | +0.038 |
ests | +0.038 |
Neg logit contributions | |
glad | -0.058 |
pleased | -0.054 |
impressed | -0.051 |
lucky | -0.050 |
okay | -0.050 |
Token | Lily |
Token ID | 20037 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
bons | +0.028 |
talk | +0.025 |
growth | +0.025 |
bars | +0.023 |
ests | +0.023 |
Neg logit contributions | |
glad | -0.029 |
pleased | -0.027 |
okay | -0.025 |
impressed | -0.025 |
sorry | -0.025 |
Token | hugged |
Token ID | 41130 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
bons | +0.046 |
growth | +0.039 |
talk | +0.039 |
ests | +0.037 |
ales | +0.036 |
Neg logit contributions | |
glad | -0.046 |
pleased | -0.043 |
impressed | -0.040 |
okay | -0.040 |
lucky | -0.040 |
Token | each |
Token ID | 1123 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
bons | +0.157 |
growth | +0.143 |
boards | +0.140 |
talk | +0.139 |
board | +0.138 |
Neg logit contributions | |
glad | -0.115 |
sorry | -0.111 |
okay | -0.108 |
happy | -0.104 |
sad | -0.099 |
Token | other |
Token ID | 584 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
bons | +0.084 |
talk | +0.072 |
growth | +0.072 |
ests | +0.071 |
ales | +0.066 |
Neg logit contributions | |
glad | -0.086 |
pleased | -0.077 |
lucky | -0.075 |
okay | -0.072 |
happy | -0.071 |
Token | and |
Token ID | 290 |
Feature activation | -9.87e-01 |
Pos logit contributions | |
bons | +0.048 |
growth | +0.043 |
talk | +0.042 |
ests | +0.040 |
bars | +0.039 |
Neg logit contributions | |
glad | -0.051 |
pleased | -0.047 |
okay | -0.044 |
lucky | -0.044 |
sorry | -0.044 |
Token | thanked |
Token ID | 26280 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
bons | +0.060 |
growth | +0.054 |
talk | +0.053 |
ales | +0.050 |
bars | +0.050 |
Neg logit contributions | |
glad | -0.054 |
pleased | -0.050 |
sorry | -0.045 |
impressed | -0.045 |
okay | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
bons | +0.155 |
talk | +0.145 |
growth | +0.143 |
ests | +0.140 |
board | +0.136 |
Neg logit contributions | |
glad | -0.108 |
sorry | -0.102 |
okay | -0.098 |
happy | -0.091 |
pleased | -0.088 |